Papers

Error propagation is a general but crucial problem in online semi-supervised video object segmentation. We aim to suppress error propagation through a correction mechanism with high reliability. The key insight is to disentangle the correction from the conventional mask propagation process with reliable cues. We introduce two modulators, propagation and correction modulators, to separately perform channel-wise recalibration on the target frame embeddings according to local temporal correlations and reliable references respectively. Specifically, we assemble the modulators with a cascaded propagation-correction scheme. This avoids overriding the effects of the reliable correction modulator by the propagation modulator. Although the reference frame with the ground truth label provides reliable cues, it could be very different from the target frame and introduce uncertain or incomplete correlations. We augment the reference cues by supplementing reliable feature patches to a maintained pool, thus offering more comprehensive and expressive object representations to the modulators. In addition, a reliability filter is designed to retrieve reliable patches and pass them in subsequent frames. Our model achieves state-of-the-art performance on YouTube-VOS18, YouTube-VOS19 and DAVIS17-Val/Test benchmarks. Extensive experiments demonstrate that the correction mechanism provides considerable performance gain by fully utilizing reliable guidance.

poster session 5 @ red 1, poster session 9 @ red 1, oral session 5 @ red 1, poster session 5, poster session 9, oral session 5

Automatic diagnosis has attracted increasing attention but remains challenging due to multi-step reasoning. Recent works usually address it by reinforcement learning methods. However, these methods show low efficiency and require task-specific reward functions. Considering the conversation between doctor and patient allows doctors to probe for symptoms and make diagnoses, the diagnosis process can be naturally seen as the generation of a sequence including symptoms and diagnoses. Inspired by this, we reformulate automatic diagnosis as a symptoms Sequence Generation (SG) task and propose a simple but effective automatic Diagnosis model based on Transformer (Diaformer). We firstly design the symptom attention framework to learn the generation of symptom inquiry and the disease diagnosis. To alleviate the discrepancy between sequential generation and disorder of implicit symptoms, we further design three orderless training mechanisms. Experiments on three public datasets show that our model outperforms baselines on disease diagnosis by 1%, 6% and 11.5% with the highest training efficiency. Detailed analysis on symptom inquiry prediction demonstrates that the potential of applying symptoms sequence generation for automatic diagnosis.

poster session 5 @ blue 4, poster session 7 @ blue 4, poster session 5, poster session 7

We consider the single image super-resolution (SISR) problem, where a high-resolution (HR) image is generated based on a low-resolution (LR) input. Recently, generative adversarial networks (GANs) become popular to hallucinate details. Most methods along this line rely on a predefined single-LR-single-HR mapping, which is not flexible enough for the ill-posed SISR task. Also, GAN-generated fake details may often undermine the realism of the whole image. We address these issues by proposing best-buddy GANs (Beby-GAN) for rich-detail SISR. Relaxing the rigid one-to-one constraint, we allow the estimated patches to dynamically seek trustworthy surrogates of supervision during training, which is beneficial to producing more reasonable details. Besides, we propose a region-aware adversarial learning strategy that directs our model to focus on generating details for textured areas adaptively. Extensive experiments justify the effectiveness of our method. An ultra-high-resolution 4K dataset is also constructed to facilitate future super-resolution research.

poster session 5 @ red 2, poster session 12 @ red 2, oral session 12 @ red 2, poster session 5, poster session 12, oral session 12

Recent work has shown that current text classification models are vulnerable to small adversarial perturbation to inputs, and adversarial training that re-trains the models with the support of adversarial examples is the most popular way to alleviate the impact of the perturbation. However, current adversarial training methods have two principal problems: worse model generalization and ineffective defending against other text attacks. In this paper, we propose a \textbf{K}eyword-bias-aware \textbf{A}dversarial \textbf{T}ext \textbf{G}eneration model (KATG) that implicitly generates adversarial sentences using a generator-discriminator structure. Instead of using a benign sentence to generate an adversarial sentence, the KATG model utilizes extra multiple benign sentences (namely prior sentences) to guide adversarial sentence generation. Furthermore, to cover more perturbation used in existing attacks, a keyword-bias-aware sampling is proposed to select sentences containing biased words as prior sentences. Besides, to effectively utilize prior sentences, a generative flow mechanism is proposed to construct latent semantic space and learn a latent representation for the prior sentences. Experiments demonstrate that adversarial sentences generated by our KATG model can strengthen the victim model's robustness and generalization.

poster session 5 @ red 5, poster session 12 @ red 5, poster session 5, poster session 12

Detecting rumors on social media has become particular important due to the rapid dissemination and adverse impacts on our lives. Though a set of rumor detection models have exploited the message propagation structural or temporal information, they seldom model them altogether to enjoy the best of both worlds. Moreover, the dynamics of knowledge information associated with the comments are not involved, either. To this end, we propose a novel Dual-Dynamic Graph Convolutional Networks, termed as DDGCN, which can model the dynamics of messages in propagation as well as the dynamics of the background knowledge from Knowledge graphs in one unified framework. Specifically, two Graph Convolutional Networks are adopted to capture the above two types of structure information at different time stages, which are then combined with a temporal fusing unit. This allows for learning the dynamic event representations in a more fine-grained manner, and incrementally aggregating them to capture the cascading effect for better rumor detection. Extensive experiments on two public real-world datasets demonstrate that our proposal yields significant improvements compared to strong baselines and can detect rumors at early stages.

poster session 5 @ blue 5, poster session 9 @ blue 5, poster session 5, poster session 9

According to the official adult literacy report conducted in 24 highly-developed countries, more than 50% adults, on average, can only understand basic vocabulary, short sentences, and basic syntactic constructions. Everyday information found in news articles is thus inaccessible to many people, impeding their social inclusion and informed decision-making. Systems for automatic sentence simplification aim to provide scalable solution to this problem. In this paper, we propose new state-of-the-art sentence simplification systems for English and Spanish, and specifications for expert evaluation that are in accordance with well-established easy-to-read guidelines. We conduct expert evaluation of our new systems and the previous state-of-the-art systems for English and Spanish, and discuss strengths and weaknesses of each of them. Finally, we draw conclusions about the capabilities of the state-of-the-art sentence simplification systems and give some directions for future research.

poster session 2 @ red 6, poster session 5 @ red 6, poster session 2, poster session 5

We study the problem of predicting the influence of a user in spreading fake (or real) news on social media. We propose a new model to address this problem which takes into account both user and tweet characteristics. We show that our model achieves an F1 score of 0.853, resp. 0.931, at predicting the influence of fake, resp. real, news spreaders, and outperforms existing baselines. We also investigate important features at predicting the influence of real vs. fake news spreaders.

poster session 5 @ blue 4, poster session 12 @ blue 4, poster session 5, poster session 12

Aspect detection aims to identify aspects of reviews and is an essential up-stream task of opinion mining and so on. However, existing weakly-supervised methods suffer from lacking the ability of identifying implicit aspects with infrequent aspect terms and "Misc" aspects. To tackle these problems, we propose to enhance the representation of segment with external knowledge by a weakly-supervised method. Experiments demonstrate the effectiveness of our model and the improvement by incorporating external knowledge.

poster session 5 @ blue 5, poster session 9 @ blue 5, poster session 5, poster session 9

Spiking Neural Networks (SNNs) have recently attracted enormous research interest since their event-driven and brain-inspired structure enables low-power computation. In image recognition tasks, the best results are achieved by SNN so far utilizing ANN-SNN conversion methods that replace activation functions in artificial neural networks~(ANNs) with integrate-and-fire neurons. Compared to source ANNs, converted SNNs usually suffer from accuracy loss and require a considerable number of time steps to achieve competitive accuracy. We find that the performance degradation of converted SNN stems from the fact that the information capacity of spike trains in transferred networks is smaller than that of activation values in source ANN, resulting in less information being passed during SNN inference. To better correlate ANN and SNN for better performance, we propose a conversion framework to mitigate the gap between the activation value of source ANN and the generated spike train of target SNN. The conversion framework originates from exploring an identical relation in the conversion and exploits temporal separation scheme and novel neuron model for the relation to hold. We demonstrate almost lossless ANN-SNN conversion using SpikeConverter for VGG-16, ResNet-20/34, and MobileNet-v2 SNNs on challenging datasets including CIFAR-10, CIFAR-100, and ImageNet. Our results also show that SpikeConverter achieves the abovementioned accuracy across different network architectures and datasets using $32\times \sim 512\times$ fewer inference time-steps than state-of-the-art ANN-SNN conversion methods.

poster session 5 @ blue 4, poster session 7 @ blue 4, oral session 5 @ blue 4, poster session 5, poster session 7, oral session 5

Air pollution is a global problem and severely impacts human health. Fine-grained air quality (AQ) monitoring is important in mitigating air pollution. However, existing AQ station deployments are sparse. Conventional interpolation techniques fail to learn the complex AQ phenomena. Physics-based models require domain knowledge and pollution source data for AQ modeling. In this work, we propose a Gaussian processes based approach for estimating AQ. The important features of our approach are: a) a non-stationary (NS) kernel to allow input depended smoothness of fit; b) a Hamming distance-based kernel for categorical features; and c) a locally periodic kernel to capture temporal periodicity. We leverage batch-wise training to scale our approach to a large amount of data. Our approach outperforms the conventional baselines and a state-of-the-art neural attention-based approach.

poster session 2 @ red 6, poster session 5 @ red 6, poster session 2, poster session 5

In many real-world problems, first-order (FO) derivative evaluations are too expensive or even inaccessible. For solving these problems, zeroth-order (ZO) methods that only need function evaluations are often more efficient than FO methods or sometimes the only options. In this paper, we propose a novel zeroth-order inexact augmented Lagrangian method (ZO-iALM) to solve black-box optimization problems, which involve a composite (i.e., smooth+nonsmooth) objective and functional constraints. Under a certain regularity condition (also assumed by several existing works on FO methods), the query complexity of our ZO-iALM is $\tilde{O}(d\vareps^{-3})$ to find an $\vareps$-KKT point for problems with a nonconvex objective and nonconvex constraints, and $\tilde{O}(d\vareps^{-\frac{5}{2}})$ for nonconvex problems with convex constraints, where $d$ is the variable dimension. This appears to be the first work that develops an iALM-based ZO method for functional constrained optimization and meanwhile achieves query complexity results matching the best-known FO complexity results up to a factor of $d$. With an extensive experimental study, we show the effectiveness of our method. The applications of our method span from classical optimization problems to practical machine learning examples such as resource allocation in sensor networks and adversarial example generation.

poster session 5 @ blue 2, poster session 10 @ blue 2, oral session 10 @ blue 2, poster session 5, poster session 10, oral session 10

Data-Free Knowledge Distillation (KD) allows knowledge transfer from a trained neural network (teacher) to a more compact one (student) in the absence of original training data. Existing works use a validation set to monitor the accuracy of the student over real data and report the highest performance throughout the entire process. However, validation data may not be available at distillation time either, making it infeasible to record the student snapshot that achieved the peak accuracy. Therefore, a practical data-free KD method should be robust and ideally provide monotonically increasing student accuracy during distillation. This is challenging because the student experiences knowledge degradation due to the distribution shift of the synthetic data. A straightforward approach to overcome this issue is to store and rehearse the generated samples periodically, which increases the memory footprint and creates privacy concerns. We propose to model the distribution of the previously observed synthetic samples with a generative network. In particular, we design a Variational Autoencoder (VAE) with a training objective that is customized to learn the synthetic data representations optimally. The student is rehearsed by the generative pseudo replay technique, with samples produced by the VAE. Hence knowledge degradation can be prevented without storing any samples. Experiments on image classification benchmarks show that our method optimizes the expected value of the distilled model accuracy while eliminating the large memory overhead incurred by the sample-storing methods.

poster session 5 @ blue 3, poster session 10 @ blue 3, poster session 5, poster session 10

Human-Object Interaction (HOI) detection plays a core role in activity understanding. As a compositional learning problem (human-verb-object), studying its generalization matters. However, widely-used metric mean average precision (mAP) fails to model the compositional generalization well. Thus, we propose a novel metric, mPD (mean Performance Degradation), as a complementary of mAP to evaluate the performance gap among compositions of different objects and the same verb. Surprisingly, mPD reveals that previous methods usually generalize poorly. With mPD as a cue, we propose Object Category (OC) Immunity to boost HOI generalization. The idea is to prevent model from learning spurious object-verb correlations as a short-cut to over-fit the train set. To achieve OC-immunity, we propose an OC-immune network that decouples the inputs from OC, extracts OC-immune representations, and leverages uncertainty quantification to generalize to unseen objects. In both conventional and zero-shot experiments, our method achieves decent improvements. To fully evaluate the generalization, we design a new and more difficult benchmark, on which we present significant advantage. The code is available at https://github.com/Foruck/OC-Immunity.

poster session 5 @ red 1, poster session 9 @ red 1, poster session 5, poster session 9

Neural network controllers have become popular in control tasks thanks to their flexibility and expressivity. Stability is a crucial property for safety-critical dynamical systems, while stabilization of partially observed systems, in many cases, requires controllers to retain and process long-term memories of the past. We consider the important class of recurrent neural networks (RNN) as dynamic controllers for nonlinear uncertain partially-observed systems, and derive convex stability conditions based on integral quadratic constraints, S-lemma and sequential convexification. To ensure stability during the learning and control process, we propose a projected policy gradient method that iteratively enforces the stability conditions in the reparametrized space taking advantage of mild additional information on system dynamics. Numerical experiments show that our method learns stabilizing controllers with fewer samples and achieves higher final performance compared with policy gradient.

poster session 5 @ red 3, poster session 12 @ red 3, poster session 5, poster session 12

Learning from one's mistakes is an effective human learning technique where the learners focus more on the topics where mistakes were made, so as to deepen their understanding. In this paper, we investigate if this human learning strategy can be applied in machine learning. We propose a novel machine learning method called Learning From Mistakes (LFM), wherein the learner improves its ability to learn by focusing more on the mistakes during revision. We formulate LFM as a three-stage optimization problem: 1) learner learns; 2) learner re-learns focusing on the mistakes, and; 3) learner validates its learning. We develop an efficient algorithm to solve the LFM problem. We apply the LFM framework to neural architecture search on CIFAR-10, CIFAR-100, and Imagenet. Experimental results strongly demonstrate the effectiveness of our model.

poster session 5 @ red 4, poster session 10 @ red 4, poster session 5, poster session 10

Colorizing a grayscale image is inherently an ill-posed problem with multi-modal uncertainty. Language-based colorization offers a natural way of interaction to reduce such uncertainty via a user-provided caption. However, the color-object coupling and mismatch issues make the mapping from word to color difficult. In this paper, we propose L-CoDe, a Language-based Colorization network using color-object Decoupled conditions. A predictor for object-color corresponding matrix (OCCM) and a novel attention transfer module (ATM) are introduced to solve the color-object coupling problem. To deal with color-object mismatch that results in incorrect color-object correspondence, we adopt a soft-gated injection module (SIM). We further present a new dataset containing annotated color-object pairs to provide supervisory signals for resolving the coupling problem. Experimental results show that our approach outperforms state-of-the-art methods conditioned on captions.

poster session 5 @ red 2, poster session 12 @ red 2, poster session 5, poster session 12

Categorical data is common and, however, special in that its possible values exist only on a nominal scale so that many statistical operations such as mean, variance, and covariance become not applicable. Following the basic idea of the neighbour correlation coefficient (nCor), in this study, we propose a new measure named the categorical nCor (CnCor) to examine the association between categorical variables through using indicator functions to reform the distance metric and product-moment correlation coefficient. The proposed measure is easy to compute, and enables a direct test of statistical dependence without the need of converting the qualitative variables to quantitative ones. Compare to previous approaches, it is much more robust and effective in dealing with multi-categorical target variables especially when highly nonlinear relationships occurs in the multivariate case. We also applied the CnCor to implementing feature selection by the scheme of backward elimination. Finally, extensive experiments performed on both synthetic and real-world datasets are conducted to demonstrate the outstanding performance of the proposed methods, and draw comparisons with state-of-the-art association measures and feature selection algorithms.

poster session 5 @ blue 1, poster session 10 @ blue 1, poster session 5, poster session 10

A growing body of work in game theory extends the traditional Stackelberg game to settings with one leader and multiple followers who play a Nash equilibrium. Standard approaches for computing equilibria in these games reformulate the followers' best response as constraints in the leader's optimization problem. These reformulation approaches can sometimes be effective, but make limiting assumptions on the followers' objectives and the equilibrium reached by followers, e.g., uniqueness, optimism, or pessimism. To overcome these limitations, we run gradient descent to update the leader's strategy by differentiating through the equilibrium reached by followers. Our approach generalizes to any stochastic equilibrium selection procedure that chooses from multiple equilibria, where we compute the stochastic gradient by back-propagating through a sampled Nash equilibrium using the solution to a partial differential equation to establish the unbiasedness of the stochastic gradient. Using the unbiased gradient estimate, we implement the gradient-based approach to solve three Stackelberg problems with multiple followers. Our approach consistently outperforms existing baselines to achieve higher utility for the leader.

poster session 5 @ blue 6, poster session 8 @ blue 6, poster session 5, poster session 8

Recent works in artificial intelligence fairness attempt to mitigate discrimination by proposing constrained optimization programs that achieve parity for some fairness statistic. Most assume availability of the class label, which is impractical in many real-world applications such as precision medicine, actuarial analysis and recidivism prediction. Here we consider fairness in longitudinal right-censored environments, where the time to event might be unknown, resulting in censorship of the class label and inapplicability of existing fairness studies. We devise applicable fairness measures, propose a debiasing algorithm to bridge fairness with and without censorship, and provide necessary theoretical constructs for these important and socially-sensitive tasks. Our experiments on four censored datasets confirm the utility of our approach.

poster session 5 @ red 6, poster session 12 @ red 6, poster session 5, poster session 12

Slides are broadly used to present the research works and there are several studies on the problem of automatic slides generation. However, the lack of dataset hinders further research. In this paper, we construct a benchmark dataset for the problem of slides generation from scholarly papers. The dataset is fine-grained and consists of aligned pairs of single slide and specific region of a paper. Then we deploy several baseline models and conduct preliminary experiments. The results show that this task is challenging and awaits more exploration. The dataset and code will be released.

poster session 5 @ red 5, poster session 12 @ red 5, poster session 5, poster session 12

We pose and study the problem of planning in Markov decision processes (MDPs), subject to participation constraints as studied in mechanism design. In this problem, a planner must work with a self-interested agent on a given MDP. Each action in the MDP provides an immediate reward to the planner and a (possibly different) reward to the agent. The agent has no control in choosing the actions, but has the option to end the entire process at any time. The goal of the planner is to find a policy that maximizes her cumulative reward, taking into consideration the agent's ability to terminate. Our main result is a fully polynomial-time approximation scheme for this problem. En route, we present polynomial-time algorithms for computing (exact) optimal policies for important special cases of this problem, including when the time horizon is constant, or when the MDP exhibits a "definitive decisions" property. We illustrate our algorithms with two different game-theoretic applications: the problem of assigning rides in ride-sharing and the problem of designing screening policies. Our results imply efficient algorithms for computing (approximately) optimal policies in both these applications.

poster session 5 @ blue 6, poster session 8 @ blue 6, oral session 5 @ blue 6, poster session 5, poster session 8, oral session 5

After criminal recidivism or hiring machine learning mod-els have inflicted harm, participatory machine learning meth-ods are often used as a corrective positioning. However, lit-tle guidance exists on how to develop participatory machinelearning models throughout stages of the machine learningdevelopment life-cycle. Here we demonstrate how to co-design and partner with community groups, in the specificcase of feminicide data activism. We co-designed and piloteda machine learning model for the detection of media arti-cles about feminicide. This provides a feminist perspectiveon practicing participatory methods in a co-creation mind-set for the real-world scenario of monitoring violence againstwomen.

poster session 3 @ red 6, poster session 5 @ red 6, poster session 3, poster session 5

We revisit Semantic Scene Completion (SSC), a useful task to predict the semantic and occupancy representation of 3D scenes, in this paper. A number of methods for this task are always based on voxelized scene representations. Although voxel representations keep local structures of the scene, these methods suffer from heavy computation redundancy due to the existence of visible empty voxels when the network goes deeper. To address this dilemma, we propose our novel point-voxel aggregation network for this task. We first transfer the voxelized scenes to point clouds by removing these visible empty voxels and adopt a deep point stream to capture semantic information from the scene efficiently. Meanwhile, a light-weight voxel stream containing only two 3D convolution layers preserves local structures of the voxelized scenes. Furthermore, we design an anisotropic voxel aggregation operator to fuse the structure details from the voxel stream into the point stream, and a semantic-aware propagation module to enhance the up-sampling process in the point stream by semantic labels. We demonstrate that our model surpasses state-of-the-arts on two benchmarks by a large margin, with only the depth images as input.

poster session 5 @ red 3, poster session 12 @ red 3, poster session 5, poster session 12

Model-Agnostic Meta-Learning (MAML), a popular gradient-based meta-learning framework, assumes that the contribution of each task or instance to the meta-learner is equal.Hence, it fails to address the domain shift between base and novel classes in few-shot learning. In this work, we propose a novel robust meta-learning algorithm, NESTEDMAML, which learns to assign weights to training tasks or instances. We con-sider weights as hyper-parameters and iteratively optimize them using a small set of validation tasks set in a nested bi-level optimization approach (in contrast to the standard bi-level optimization in MAML). We then applyNESTED-MAMLin the meta-training stage, which involves (1) several tasks sampled from a distribution different from the meta-test task distribution, or (2) some data samples with noisy labels.Extensive experiments on synthetic and real-world datasets demonstrate that NESTEDMAML efficiently mitigates the effects of ”unwanted” tasks or instances, leading to significant improvement over the state-of-the-art robust meta-learning methods.

poster session 5 @ red 4, poster session 10 @ red 4, poster session 5, poster session 10

Despite the many recent practical and theoretical breakthroughs in computational game theory, equilibrium finding in extensive-form team games remains a significant challenge. While NP-hard in the worst case, there are provably efficient algorithms for certain families of team game. In particular, if the game has common external information, also known as A-loss recall---informally, actions played by non-team members (i.e., the opposing team or nature) are either unknown to the entire team, or common knowledge within the team---then polynomial-time algorithms exist (Kaneko and Kline 1995). In this paper, we devise a completely new algorithm for solving team games. It uses a tree decomposition of the constraint system representing each team's strategy to reduce the number and degree of constraints required for correctness (tightness of the mathematical program). Our approach has the bags of the tree decomposition correspond to team-public states---that is, minimal sets of nodes (that is, states of the team) such that, upon reaching the set, it is common knowledge among the players on the team that the set has been reached. Our algorithm reduces the problem of solving team games to a linear program with at most O(NW^{w+1}) nonzero entries in the constraint matrix, where N is the size of the game tree, w is a parameter that depends on the amount of uncommon external information, and W is the treewidth of the tree decomposition. In public-action games, our program size is bounded by the tighter 2^{O(nt)}N for teams of n players with t types each. Our algorithm is based on a new way to write a custom, concise tree decomposition, and its fast run time does not assume that the decomposition has small treewidth. Since our algorithm describes the polytope of correlated strategies directly, we get equilibrium finding in correlated strategies for free---instead of, say, having to run a double oracle algorithm. We show via experiments on a standard suite of games that our algorithm achieves state-of-the-art performance on all benchmark game classes except one. We also present, to our knowledge, the first experiments for this setting where both teams have more than one member.

poster session 5 @ blue 6, poster session 8 @ blue 6, poster session 5, poster session 8

Model compression is very important for the efficient deployment of deep neural network (DNN) models on resource-constrained devices. Among various model compression approaches, high-order tensor decomposition is particularly attractive and useful because the decomposed model is very small and fully structured. For this category of approaches, tensor ranks are the most important hyper-parameters that directly determine the architecture and task performance of the compressed DNN models. However, as an NP-hard problem, selecting optimal tensor ranks under the desired budget is very challenging and the state-of-the-art studies suffer from unsatisfied compression performance and timing-consuming search procedures. To systematically address this fundamental problem, in this paper we propose BATUDE, a Budget-Aware TUcker DEcomposition-based compression approach that can efficiently calculate optimal tensor ranks via one-shot training. By integrating the rank selecting procedure to the DNN training process with a specified compression budget, the tensor ranks of the DNN models are learned from the data and thereby bringing very significant improvement on both compression ratio and classification accuracy for the compressed models. The experimental results on ImageNet dataset show that our method enjoys 0.33% top-5 higher accuracy with 2.52X less computational cost as compared to the uncompressed ResNet-18 model. For ResNet-50, the proposed approach enables 0.37% and 0.55% top-5 accuracy increase with 2.97X and 2.04X computational cost reduction, respectively, over the uncompressed model.

poster session 5 @ blue 3, poster session 10 @ blue 3, poster session 5, poster session 10

It has been crucial to leverage the rich information of multiple modalities in many tasks. Existing works have tried to design multi-modal networks with descent multi-modal fusion modules. Instead, we focus on improving generalization capability of multi-modal networks, especially the fusion module. Viewing the multi-modal data as \textit{different projections of information}, we first observe that bad projection can cause poor generalization behaviors of multi-modal networks. Then, motivated by well-generalized network's \textit{low sensitivity to perturbation}, we propose a novel multi-modal training method, \textit{multi-head modularization (MHM)}. We modularize a multi-modal network as a series of uni-modal embedding, multi-modal embedding, and task-specific head modules. Also, for training, we exploit multiple head modules learned with different datasets, swapping each other. From this, we can make the multi-modal embedding module robust to all the heads with different generalization behaviors. In testing phase, we select one of the head modules not to increase the computational cost. Owing to the perturbation of head modules, though including one selected head, the deployed network is more well-generalized compared to the simply end-to-end learned. We verify the effectiveness of MHM on various multi-modal tasks. We use the state-of-the-art methods as baselines, and show notable performance gain for all the baselines.

poster session 5 @ red 4, poster session 10 @ red 4, poster session 5, poster session 10

We investigate the capability of a transformer pretrained on natural language to generalize to other modalities with minimal finetuning -- in particular, without finetuning of the self-attention and feedforward layers of the residual blocks. We consider such a model, which we call a Frozen Pretrained Transformer (FPT), and study finetuning it on a variety of sequence classification tasks spanning numerical computation, vision, and protein fold prediction. In contrast to prior works which investigate finetuning on the same modality as the pretraining dataset, we show that pretraining on natural language can improve performance and compute efficiency on non-language downstream tasks. Additionally, we perform an analysis of the architecture, comparing the performance of a random initialized transformer to a random LSTM. Combining the two insights, we find language-pretrained transformers can obtain strong performance on a variety of non-language tasks.

poster session 5 @ red 4, poster session 10 @ red 4, oral session 10 @ red 4, poster session 5, poster session 10, oral session 10

We study the problem of releasing the set union of users' items subject to differential privacy. Previous approaches consider only the set of items for each user as the input. We propose incorporating the item frequency, which is typically available in set union problems, to boost the utility of private mechanisms. However, using the global item frequency over all users would largely increase privacy loss. We propose to use the local item frequency of each user to approximate the global item frequency without incurring additional privacy loss. Local item frequency allows us to design greedy set union mechanisms that are differentially private, which is impossible for previous greedy proposals. Moreover, while all previous works have to use uniform sampling to limit the number of items each user would contribute to, our construction eliminates the sampling step completely and allows our mechanisms to consider all of the users' items. Finally, we propose to transfer the knowledge of the global item frequency from a public dataset into our mechanism, which further boosts utility even when the public and private datasets are from different domains. We evaluate the proposed methods on multiple real-life datasets.

poster session 5 @ blue 2, poster session 10 @ blue 2, poster session 5, poster session 10

Expert finding, a popular service provided by many onlinewebsites such as Expertise Finder, LinkedIn, and AMiner,is beneficial to seeking candidate qualifications, consultants,and collaborators. However, its quality is suffered from lackof ample sources of expert information. This paper employsAMiner as the basis with an aim at linking any external experts to the counterparts on AMiner. As it is infeasible to acquire sufficient linkages from arbitrary external sources, weexplore the problem of zero-shot expert linking. In this paper,we propose CODE, which first pre-trains an expert linkingmodel by contrastive learning on AMiner such that it can capture the representation and matching patterns of experts without supervised signals, then it is fine-tuned between AMinerand external sources to enhance the model’s transferabilityin an adversarial manner. For evaluation, we first design twointrinsic tasks, author identification and paper clustering, tovalidate the representation and matching capability endowedby contrastive learning. Then the final external expert linkingperformance on two genres of external sources also impliesthe superiority of adversarial fine-tuning method. Additionally, we show the online deployment of CODE, and continuously improve its online performance via active learning.

poster session 2 @ red 6, poster session 5 @ red 6, poster session 2, poster session 5

Network alignment (NA) is the task of finding the correspondence of nodes between two networks. Since most existing NA methods have attempted to discover every node pair at once, they may fail to utilize node pairs that have strong consistency across different networks in the NA task. To tackle this challenge, we propose Grad-Align, a new NA method that gradually discovers node pairs by making full use of either node pairs exhibiting strong consistency or prior matching information. Specifically, the proposed method gradually aligns nodes based on both the similarity of embeddings generated using graph neural networks (GNNs) and the Tversky similarity, which is an asymmetric set similarity using the Tversky index applicable to networks with different scales. Experimental evaluation demonstrates that Grad-Align consistently outperforms state-of-the-art NA methods in terms of the alignment accuracy. Our source code is available at https://github.com/jindeok/Grad-Align.

poster session 5 @ blue 3, poster session 10 @ blue 3, poster session 5, poster session 10

For a monocular camera-based navigation system, if we could effectively explore scene geometric cues from RGB images, the geometry information will significantly facilitate the efficiency of the navigation system. Motivated by this, we propose a highly efficient point-goal navigation framework, dubbed Geo-Nav. In a nutshell, our Geo-Nav consists of two parts: a visual perception part and a navigation part. In the visual perception part, we firstly propose a Self-supervised Depth Estimation network (SDE) specially tailored for the monocular camera-based navigation agent. Our SDE learns a mapping from an RGB input image to its corresponding depth image by exploring scene geometric constraints in a self-consistency manner. Then, in order to achieve a representative visual representation from the RGB inputs and learned depth images, we propose a Cross-modality Pyramid Fusion module (CPF). Concretely, our CPF computes a patch-wise cross-modality correlation between different modal features and exploits the correlation to fuse and enhance features at each scale. Thanks to the patch-wise nature of our CPF, we can fuse feature maps at high resolution, allowing our visual network to perceive more image details. In the navigation part, our extracted visual representations are fed to a navigation policy network to learn how to map the visual representations to agent actions effectively. Extensive experiments on a widely-used multiple-room environment Gibson demonstrate that Geo-Nav outperforms the state-of-the-art in terms of efficiency and effectiveness.

poster session 5 @ red 3, poster session 12 @ red 3, poster session 5, poster session 12

Low-light image enhancement aims to improve an image's visibility while keeping its visual naturalness. Different from existing methods, which tend to accomplish the relighting task directly, we investigate the intrinsic degradation and relight the low-light image while refining the details and color in two steps. Inspired by the color image formulation (diffuse illumination color plus environment illumination color), we first estimate the degradation from low-light inputs to simulate the distortion of environment illumination color, and then refine the content to recover the loss of diffuse illumination color. To this end, we propose a novel Degradation-to-Refinement Generation Network (DRGN). Its distinctive features can be summarized as 1) A novel two-step generation network for degradation learning and content refinement. It is not only superior to one-step methods, but also capable of synthesizing sufficient paired samples to benefit the model training; 2) A multi-resolution fusion network to represent the target information (degradation or contents) in a multi-scale cooperative manner, which is more effective to address the complex unmixing problems. Extensive experiments on both the enhancement task and the joint detection task have verified the effectiveness and efficiency of our proposed method, surpassing the SOTA by 1.59dB on average and 3.18\% in mAP on the ExDark dataset. The code will be available soon.

poster session 5 @ red 2, poster session 12 @ red 2, poster session 5, poster session 12

Bilinear pooling has achieved excellent performance in fine-grained recognition tasks. Nevertheless, high-dimensional bilinear features suffer from over-fitting and inefficiency. To alleviate these issues, compact bilinear pooling (CBP) methods were developed to generate low-dimensional features. Although the low-dimensional features from existing CBP methods enable high efficiency in subsequent classification, CBP methods themselves are inefficient. Thus, the inefficiency issue of the bilinear pooling is still unsolved. In this work, we propose an efficient compact bilinear pooling method to solve the inefficiency problem inherited in bilinear pooling thoroughly. It decomposes the huge-scale projection matrix into a two-level Kronecker product of several small-scale matrices. By exploiting the ``vec trick'' and the tensor modal product, we can obtain the compact bilinear feature through the decomposed projection matrices in a speedy manner. Systematic experiments on four public benchmarks using two backbones demonstrate the efficiency and effectiveness of the proposed method in fine-grained recognition.

poster session 5 @ red 3, poster session 12 @ red 3, poster session 5, poster session 12

Early detection of lung cancer is crucial for five-year survival of patients. Compared with the pathological analysis and CT scans, the circulating tumor DNA (ctDNA) methylation based approach is noninvasive and cost-effective, and thus is one of the most promising methods for early detection of lung cancer. Existing studies on ctDNA methylation data measure the methylation level of each region with a predefined metric, ignoring the positions of methylated CpG sites and methylation patterns, thus are not able to capture the early cancer signals. In this paper, we propose a blood-based lung cancer detection method, and present the first ever study to represent methylation regions by continuous vectors. Specifically, we propose DeepMeth to regard each region as a one-channel image and develop an auto-encoder model to learn its representation. For each ctDNA methylation sample, DeepMeth achieves its representation via concatenating the region vectors. We evaluate DeepMeth on a multicenter clinical dataset collected from 14 hospitals. The experiments show that DeepMeth achieves about 5%-8% improvements compared with the baselines in terms of Area Under the Curve (AUC). Moreover, the experiments also demonstrate that DeepMeth can be combined with traditional scalar metrics to enhance the diagnostic power of ctDNA methylation classifiers. DeepMeth has been clinically deployed and applied to 450 patients from 94 hospitals nationally since April 2020.

poster session 5 @ red 6, poster session 12 @ red 6, poster session 5, poster session 12

Active learning algorithms are effective in identifying the salient and exemplar samples from large amounts of unlabeled data. This tremendously reduces the human annotation effort in inducing a machine learning model as only a few samples, which are identified by the algorithm, need to be labeled manually. In problem domains like text mining and video classification, human oracles peruse the data instances incrementally to derive an opinion about their class labels (such as reading a movie review progressively to assess its sentiment). In such applications, it is not necessary for the human oracles to review an unlabeled sample end-to-end in order to provide a label; it may be more efficient to identify an optimal subinstance size (percentage of the sample from the start) for each unlabeled sample, and request the human annotator to label the sample by analyzing only the subinstance, instead of the whole data sample. In this paper, we propose a novel framework to address this challenging problem, in an effort to further reduce the labeling burden on the human oracles and utilize the available labeling budget more efficiently. We pose the sample and subinstance size selection as a constrained optimization problem and derive a linear programming relaxation to select a batch of exemplar samples, together with the optimal subinstance size of each, which can potentially augment maximal information to the underlying classification model. Our extensive empirical studies on six challenging datasets from the text mining domain corroborate the practical usefulness of our framework over competing baselines.

poster session 5 @ red 4, poster session 10 @ red 4, oral session 5 @ red 4, poster session 5, poster session 10, oral session 5

RGB–thermal scene parsing has recently attracted increasing research interest in the field of computer vision. However, most existing methods fail to perform good boundary extraction for prediction maps and cannot fully use high-level features. In addition, these methods simply fuse the features from RGB and thermal modalities but are unable to obtain comprehensive fused features. To address these problems, we propose an edge-aware guidance fusion network (EGFNet) for RGB–thermal scene parsing. First, we introduce a prior edge map generated using the RGB and thermal images to capture detailed information in the prediction map and then embed the prior edge information in the feature maps. To effectively fuse the RGB and thermal information, we propose a multimodal fusion module that guarantees adequate cross-modal fusion. Considering the importance of high-level semantic information, we propose a global information module and a semantic information module to extract rich semantic information from the high-level features. For decoding, we use simple elementwise addition for cascaded feature fusion. Finally, to improve the parsing accuracy, we apply multitask deep supervision to the semantic and boundary maps. Extensive experiments were performed on benchmark datasets to demonstrate the effectiveness of the proposed EGFNet and its superior performance compared with state-of-the-art methods. The code and results can be found at https://github.com/ShaohuaDong2021/EGFNet.

poster session 5 @ red 3, poster session 12 @ red 3, poster session 5, poster session 12

Single-cell RNA sequencing (scRNA-seq) provides high-throughput information about the genome-wide gene expression levels at the single-cell resolution, bringing a precise understanding on the transcriptome of individual cells. Unfortunately, the rapidly growing scRNA-seq data and the prevalence of dropout events pose substantial challenges for cell type annotation. Here, we propose a single-cell model-based deep graph embedding clustering (scTAG) method, which simultaneously learns cell–cell topology representations and identifies cell clusters based on deep graph convolutional network. scTAG integrates the zero-inflated negative binomial (ZINB) model into a topology adaptive graph convolutional autoencoder to learn the low-dimensional latent representation and adopts Kullback–Leibler (KL) divergence for the clustering tasks. By simultaneously optimizing the clustering loss, ZINB loss, and the cell graph reconstruction loss, scTAG jointly optimizes cluster label assignment and feature learning with the topological structures preserved in an end-to-end manner. Extensive experiments on 16 single-cell RNA-seq datasets from diverse yet representative single-cell sequencing platforms demonstrate the superiority of scTAG over various state-of-the-art clustering methods. The code is available at https://github.com/Philyzh8/scTAG.

poster session 5 @ blue 4, poster session 7 @ blue 4, oral session 7 @ blue 4, poster session 5, poster session 7, oral session 7

In many visual systems, visual tracking often bases on RGB image sequences, in which some targets are invalid in low-light conditions, and tracking performance is thus affected significantly. Introducing other modalities such as depth and infrared data is an effective way to handle imaging limitations of individual sources, but multi-modal imaging platforms usually require elaborate designs and cannot be applied in many real-world applications at present. Near-infrared (NIR) imaging becomes an essential part of many surveillance cameras, whose imaging is switchable between RGB and NIR based on the light intensity. These two modalities are heterogeneous with very different visual properties and thus bring big challenges for visual tracking. However, existing works have not studied this challenging problem. In this work, we address the cross-modal object tracking problem and contribute a new video dataset, including 654 cross-modal image sequences with over 481K frames in total, and the average video length is more than 735 frames. To promote the research and development of cross-modal object tracking, we propose a new algorithm, which learns the modality-aware target representation to mitigate the appearance gap between RGB and NIR modalities in the tracking process. It is plug-and-play and could thus be flexibly embedded into different tracking frameworks. Extensive experiments on the dataset are conducted, and we demonstrate the effectiveness of the proposed algorithm in two representative tracking frameworks against 19 state-of-the-art tracking methods. Dataset, code, model and results are available at https://github.com/mmic-lcl/source-code.

poster session 5 @ red 1, poster session 9 @ red 1, poster session 5, poster session 9

Unsupervised pretraining based on contrastive learning has made significant progress recently and showed comparable or even superior transfer learning performance to traditional supervised pretraining on various tasks. In this work, we first empirically investigate when and why unsupervised pretraining surpasses supervised counterparts for image classification tasks with a series of control experiments. Besides the commonly used accuracy, we further analyze the results qualitatively with the class activation maps and assess the learned representations quantitatively with the representation entropy and uniformity. Our core finding is that it is the amount of information effectively perceived by the learning model that is crucial to transfer learning, instead of absolute size of the dataset. Based on this finding, we propose Classification Activation Map guided contrastive (CAMtrast) learning which better utilizes the label supervision to strengthen supervised pretraining, by making the networks perceive more information from the training images. CAMtrast is evaluated with three fundamental visual learning tasks: image recognition, object detection, and semantic segmentation, on various public datasets. Experimental results show that our CAMtrast effectively improves the performance of supervised pretraining, and that its performance is superior to both unsupervised counterparts and a recent related work which similarly attempted improving supervised pretraining. Code will be made publicly available.

poster session 5 @ blue 3, poster session 10 @ blue 3, poster session 5, poster session 10

Multi Agent Path Finding (MAPF) is widely needed to coordinate real-world robotic systems. New approaches turn to deep learning to solve MAPF instances, primarily using reinforcement learning, which has high computational costs. We propose a supervised learning approach to solve MAPF instances using a smaller, less costly model.

poster session 2 @ blue 4, poster session 5 @ blue 4, poster session 2, poster session 5

Despite deep learning (DL) success in classification problems, DL classifiers do not provide a sound mechanism to decide when to refrain from predicting. Recent works tried to control the overall prediction risk with classification with rejection options. However, existing works overlook the different significance of different classes. We introduce Set-classifier with class-specific RIsk Bounds (SCRIB) to tackle this problem, assigning multiple labels to each example. Given the output of a black-box model on the validation set, SCRIB constructs a set-classifier that controls the class-specific prediction risks. The key idea is to reject when the set classifier returns more than one label. We validated SCRIB on several medical applications, including sleep staging on electroencephalogram(EEG) data, X-ray COVID image classification, and atrial fibrillation detection based on electrocardiogram (ECG) data.SCRIB obtained desirable class-specific risks, which are 35%-88% closer to the target risks than baseline methods.

poster session 5 @ blue 2, poster session 10 @ blue 2, poster session 5, poster session 10

The core of self-supervised learning for pre-training language models includes pre-training task design as well as appropriate data augmentation. Most data augmentations in language model pre-training are context-independent. A seminal contextualized augmentation was recently proposed in ELECTRA and achieved state-of-the-art performance by introducing an auxiliary generation network (generator) to produce contextualized data augmentation for the training of a main discrimination network (discriminator). This design, however, introduces extra computation cost of the generator and a need to adjust the relative capability between the generator and the discriminator. In this paper, we propose a self-augmentation strategy (SAS) where a single network is utilized for both regular pre-training and contextualized data augmentation for the training in later epochs. Essentially, this strategy eliminates a separate generator and uses the single network to jointly conduct two pre-training tasks with MLM (Masked Language Modeling) and RTD (Replaced Token Detection) heads. It avoids the challenge to search for an appropriate size of the generator, which is critical to the performance as evidenced in ELECTRA and its subsequent variant models. In addition, SAS is a general strategy that can be seamlessly combined with many new techniques emerging recently or in the future, such as the disentangled attention mechanism from DeBERTa. Our experiments show that SAS is able to outperform ELECTRA and other state-of-the-art models in the GLUE tasks with similar or less computation cost.

poster session 5 @ red 4, poster session 10 @ red 4, poster session 5, poster session 10

This paper describes an energy-based learning method that predicts the activities of multiple agents simultaneously. It aims to forecast both upcoming actions and paths of all agents in a scene based on their past activities, which can be jointly formulated by a probabilistic model over time. Learning this model is challenging because: 1) it has a large number of time-dependent variables that must scale with the forecast horizon and the number of agents; 2) distribution functions have to contain multiple modes in order to capture the spatio-temporal complexities of each agent's activities. To address these challenges, we put forth a novel Energy-based Learning approach for Multi-Agent activity forecasting (ELMA) to estimate this complex model via maximum log-likelihood estimation. Specifically, by sampling from a sequence of factorized marginalized multi-model distributions, ELMA generates most possible future actions efficiently. Moreover, by graph-based representations, ELMA also explicitly resolves the spatio-temporal dependencies of all agents' activities in a single pass. Our experiments on two large-scale datasets prove that ELMA outperforms recent leading studies by an obvious margin.

poster session 5 @ red 3, poster session 12 @ red 3, poster session 5, poster session 12

Multiple modalities can provide rich semantic information; and exploiting such information will normally lead to better performance compared with the single-modality counterpart. However, it is not easy to devise an effective cross-modal fusion structure due to the variations of feature dimensions and semantics, especially when the inputs even come from different sensors, as in the field of audio-visual learning. In this work, we propose SepFusion, a novel framework that can smoothly produce optimal fusion structures for visual-sound separation. The framework is composed of two components, namely the model generator and the evaluator. To construct the generator, we devise a lightweight architecture space that can adapt to different input modalities. In this way, we can easily obtain audio-visual fusion structures according to our demands. For the evaluator, we adopt the idea of neural architecture search to select superior networks effectively. This automatic process can significantly save human efforts while achieving competitive performances. Moreover, since our SepFusion provides a series of strong models, we can utilize the model family for broader applications, such as further promoting performance via model assembly, or providing suitable architectures for the separation of certain instrument classes. These potential applications further enhance the competitiveness of our approach.

poster session 5 @ red 3, poster session 12 @ red 3, poster session 5, poster session 12

Recent methods in speech and language technology pre-train very large models which are fine-tuned for specific tasks. However, the benefits of such large models are often limited to a few resource rich languages of the world. In this work, we make multiple contributions towards building ASR systems for low resource languages from the Indian subcontinent. First, we curate 16,000 hours of raw speech data for 40 Indian languages from a wide variety of domains including education, news, technology, and finance. Second, using this raw speech data we pre-train several variants of wav2vec style models for 40 Indian languages. Third, we analyze the pretrained models to find key features: code-book vectors of similar sounding phonemes are shared across languages, representations across layers are discriminative of the language family, and attention heads often pay attention within small local windows. Fourth, we fine-tune this model for downstream ASR for 9 languages and obtain state-of-the-art results on 3 public datasets, including on very low-resource languages such as Sinhala and Nepali. Our work establishes that multilingual pretraining is an effective strategy for building ASR systems for the linguistically diverse speakers of the Indian subcontinent.

poster session 5 @ red 5, poster session 12 @ red 5, oral session 12 @ red 5, poster session 5, poster session 12, oral session 12

How can we bring both privacy and energy-efficiency to a neural system on edge devices? In this paper, we propose PrivateSNN, which aims to build low-power Spiking Neural Networks (SNNs) from a pre-trained ANN model without leaking sensitive information contained in a dataset. Here, we tackle two types of leakage problems: 1) Data leakage caused when the networks access real training data during an ANN-SNN conversion process. 2) Class leakage caused when class-related features can be reconstructed from network parameters. In order to address the data leakage issue, we generate synthetic images from the pre-trained ANNs and convert ANNs to SNNs using the generated images. However, converted SNNs remain vulnerable to class leakage since the weight parameters have the same (or scaled) value with respect to ANN parameters. Therefore, we encrypt SNN weights by training SNNs with a temporal spike-based learning rule. Updating weight parameters with temporal data makes networks difficult to be interpreted in the spatial domain. We observe that the encrypted PrivateSNN eliminates data and class leakage issues with slight performance drop (less than ~1\%) and significant energy-efficiency gain (about 60x) compared to the standard ANN. We conduct extensive experiments on various datasets including CIFAR10, CIFAR100, and TinyImageNet, highlighting the importance of privacy-preserving SNN training. The code will be made available.

poster session 5 @ blue 4, poster session 7 @ blue 4, poster session 5, poster session 7

Deep graph clustering, which aims to reveal the underlying graph structure and divide the nodes into different groups, has attracted intensive attention in recent years. However, we observe that, in the process of node encoding, existing methods suffer from representation collapse which tends to map all data into a same representation. Consequently, the discriminative capability of node representations is limited, leading to unsatisfied clustering performance. To address this issue, we propose a novel self-supervised deep graph clustering method termed Dual Correlation Reduction Network (DCRN) by reducing information correlation in a dual manner. Specifically, in our method, we first design a siamese network to encode samples. Then by forcing the cross-view sample correlation matrix and cross-view feature correlation matrix to approximate two identity matrices, respectively, we reduce the information correlation in dual level, thus improve the discriminative capability of the resulting features. Moreover, in order to alleviate representation collapse caused by over-smoothing in GCN, we introduce a propagation-regularization term to enable the network to gain long-distance information with shallow network structure. Extensive experimental results on six benchmark datasets demonstrate the effectiveness of the proposed DCRN against the existing state-of-the-art methods.

poster session 5 @ red 4, poster session 10 @ red 4, poster session 5, poster session 10

Text content created by humans or language models is often stolen or misused by adversaries. Tracing text provenance can help claim the ownership of text content or identify the malicious users who distribute misleading content like machine-generated fake news. There have been some attempts to achieve this, mainly based on watermarking techniques. Specifically, traditional text watermarking methods embed watermarks by slightly altering text format like line spacing and font, which, however, are fragile to cross-media transmissions like OCR. Considering this, natural language watermarking methods represent watermarks by replacing words in original sentences with synonyms from handcrafted lexical resources (e.g., WordNet), but they do not consider the substitution’s impact on the overall sentence's meaning. Recently, a transformer-based network was proposed to embed watermarks by modifying the unobtrusive words (e.g., function words), which also impair the sentence's logical and semantic coherence. Besides, one well-trained network fails on other different types of text content. To address the limitations mentioned above, we propose a natural language watermarking scheme based on context-aware lexical substitution (LS). Specifically, we employ BERT to suggest LS candidates by inferring the semantic relatedness between the candidates and the original sentence. Based on this, a selection strategy in terms of synchronicity and substitutability is further designed to test whether a word is exactly suitable for carrying the watermark signal. Extensive experiments demonstrate that, under both objective and subjective metrics, our watermarking scheme can well preserve the semantic integrity of original sentences and has a better transferability than existing methods. Besides, the proposed LS approach outperforms the state-of-the-art approach on the Stanford Word Substitution Benchmark.

poster session 5 @ red 5, poster session 12 @ red 5, oral session 12 @ red 5, poster session 5, poster session 12, oral session 12

Targeted image retrieval has long been a challenging problem since each person has a different perception of different features leading to inconsistency among users in describing the details of a particular image. Due to this, each user needs a system personalized according to the way they have structured the image in their mind. One important application of this task is suspect identification in forensic investigations where a witness needs to identify the suspect from an existing criminal database. Existing methods require the attributes for each image or suffer from poor latency during training and inference. We propose a new approach to tackle this problem through explicit relevance feedback by introducing a novel loss function and a corresponding scoring function. For this, we leverage contrastive learning on the user feedback to generate the next set of suggested images while improving the level of personalization with each user feedback iteration.

poster session 5 @ blue 4, poster session 12 @ blue 4, poster session 5, poster session 12

This paper studies the problem of learning complex relationships between multi-labels for image recognition. Its challenges come from the rich and diverse semantic information in images. However, current methods cannot fully explore the mutual interactions among labels and do not explicitly model the label co-occurrence. To overcome these shortcomings, we innovatively propose CGML that consists of two crucial modules: 1) an image representation learning module that aims to complete the feature extraction of an image whose features are expressed in the form of primary capsules; 2) a label adaptive graph convolutional network module that leverages the popular graph convolutional networks with an adaptive label correlation graph to model label dependencies. Experiments show that our approach obviously outperforms the existing state-of-the-art methods.

poster session 5 @ red 1, poster session 9 @ red 1, poster session 5, poster session 9

We study algorithms for approximating pairwise similarity matrices that arise in natural language processing. Generally, computing a similarity matrix for $n$ data points requires $\Omega(n^2)$ similarity computations. This quadratic scaling is a significant bottleneck, especially when similarities are computed via expensive functions, e.g., via transformer models. Approximation methods reduce this quadratic complexity, often by using a small subset of exactly computed similarities to approximate the remainder of the complete pairwise similarity matrix. Significant work focuses on the efficient approximation of positive semidefinite (PSD) similarity matrices, which arise e.g., in kernel methods. However, much less is understood about indefinite (non-PSD) similarity matrices, which often arise in NLP. Motivated by the observation that many of these matrices are still somewhat close to PSD, we introduce a generalization of the popular \emph{\nystrom method} to the indefinite setting. Our algorithm can be applied to any similarity matrix and runs in sublinear time in the size of the matrix, producing a rank-$s$ approximation with just $O(ns)$ similarity computations. We show that our method, along with a simple variant of CUR decomposition, performs very well in approximating a variety of similarity matrices arising in NLP tasks. We demonstrate high accuracy of the approximated similarity matrices in the downstream tasks of document classification, sentence similarity, and cross-document coreference.

poster session 5 @ blue 1, poster session 10 @ blue 1, poster session 5, poster session 10

Machine learning methods such as deep neural networks (DNNs), despite their success across different domains, are known to often generate incorrect predictions with high confi- dence on inputs outside their training distribution. The deploy- ment of DNNs in safety-critical domains requires detection of out-of-distribution (OOD) data so that DNNs can abstain from making predictions on those. A number of methods have been recently developed for OOD detection, but there is still room for improvement. We propose the new method iDECODe, leveraging in-distribution equivariance for conformal OOD detection. It relies on a novel base non-conformity measure and a new aggregation method, used in the inductive confor- mal anomaly detection framework, thereby guaranteeing a bounded false detection rate. We demonstrate the efficacy of iDECODe by experiments on image and audio datasets, obtain- ing state-of-the-art results. We also show that iDECODe can detect adversarial examples. Code, pre-trained models, and data are available at https://github.com/ramneetk/iDECODe.

poster session 5 @ blue 5, poster session 9 @ blue 5, poster session 5, poster session 9

Recent non-local self-attention methods have proven to be effective in capturing long-range dependencies for semantic segmentation. These methods usually form a similarity map of \(\mathbb{R}^{C\times C}\) (by compressing spatial dimensions) or \(\mathbb{R}^{HW\times HW}\) (by compressing channels) to describe the feature relations along either channel or spatial dimensions, where \(C\) is the number of channels, \(H\) and \(W\) are the spatial dimensions of the input feature map. However, such practices tend to condense feature dependencies along the other dimensions, hence causing attention missing, which might lead to inferior results for small/thin categories or inconsistent segmentation inside large objects. To address this problem, we propose a new approach, namely Fully Attentional Network (FLANet), to encode both spatial and channel attentions in a single similarity map while maintaining high computational efficiency. Specifically, for each channel map, our FLANet can harvest feature responses from all other channel maps, and the associated spatial positions as well, through a novel fully attentional module. Our new method has achieved state-of-the-art performance on three challenging semantic segmentation datasets, i.e., 83.6\%, 46.99\%, and 88.5\% on the Cityscapes test set, the ADE20K validation set, and the PASCAL VOC test set, respectively.

poster session 5 @ red 3, poster session 12 @ red 3, poster session 5, poster session 12

Octave Convolution (OctConv) is a generic convolutional unit that has already achieved good performances in many computer vision tasks. Recent studies also have shown the potential of applying the OctConv in end-to-end image compression. However, considering the characteristic of image compression task, current works of OctConv may limit the performance of the image compression network due to the loss of spatial information caused by the sampling operations of inter-frequency communication. Besides, the correlation between multi-frequency latents produced by OctConv is not utilized in current architectures. In this paper, to address these problems, we propose a two-stage octave residual (ToRes) block which strips the sampling operation from OctConv to strengthen the capability of preserving useful information. Moreover, to capture the redundancy between the multi-frequency latents, a context transfer module is designed. The results show that both OctRes block and the incorporation of context transfer module help to improve the Rate-Distortion performance, and the combination of these two strategies makes our model achieve the state-of-the-art performance and outperform the latest compression standard VVC with as high as 0.5 dB increase in terms of the PSNR.

poster session 5 @ blue 5, poster session 9 @ blue 5, poster session 5, poster session 9

In this paper, we consider the problem of lacking theoretical foundation and low execution efficiency of the instance selection methods based on the $k$-nearest neighbour rule when processing large-scale data. We point out that the core idea of these methods can be explained from the perspective of Bayesian decision theory, that is, to find which instances are reducible, irreducible, and deleterious. Then, based on the percolation theory, we establish the relationship between these three types of instances and local homogeneous cluster (i.e., a set of instances with the same labels). Finally, we propose a method based on an accelerated $k$-means algorithm to construct local homogeneous clusters and remove the superfluous instances. The performance of our method is studied on extensive synthetic and benchmark data sets. Our proposed method can handle large-scale data more effectively than the state-of-the-art instance selection methods.

poster session 5 @ blue 5, poster session 9 @ blue 5, poster session 5, poster session 9

Over the past several years, legal applications of deep learning have been on the rise. However, as with other high-stakes decision making areas, the requirement for interpretability is of crucial importance. Current models utilized by legal practitioners are more of the conventional machine learning type, wherein they are inherently interpretable, yet unable to harness the performance capabilities of data-driven deep learning models. In this work, we utilize deep learning models in the area of trademark law to shed light on the issue of likelihood of confusion between trademarks. Specifically, we introduce a model-agnostic interpretable intermediate layer, a technique which proves to be effective for legal documents. Furthermore, we utilize weakly supervised learning by means of a curriculum learning strategy, effectively demonstrating the improved performance of a deep learning model. This is in contrast to the conventional models which are only able to utilize the limited number of expensive manually-annotated samples by legal experts. Although the methods presented in this work tackles the task of risk of confusion for trademarks, it is straightforward to extend them to other fields of law, or more generally, to other similar high-stakes application scenarios.

poster session 2 @ red 6, poster session 5 @ red 6, poster session 2, poster session 5

We study the problem of learning generative adversarial networks (GANs) for a rare class of an unlabeled dataset subject to a labeling budget. This problem is motivated from practical applications in domains including security (e.g., synthesizing packets for DNS amplification attacks), systems and networking (e.g., synthesizing workloads that trigger high resource usage), and machine learning (e.g., generating images from a rare class). Existing approaches are unsuitable, either requiring fully-labeled datasets or sacrificing the fidelity of the rare class for that of the common classes. We propose RareGAN, a novel synthesis of three key ideas: (1) extending conditional GANs to use labelled and unlabelled data for better generalization; (2) an active learning approach that requests the most useful labels; and (3) a weighted loss function to favor learning the rare class. We show that RareGAN achieves a better fidelity-diversity tradeoff on the rare class than prior work across different applications, budgets, rare class fractions, GAN losses, and architectures.

poster session 5 @ blue 2, poster session 10 @ blue 2, poster session 5, poster session 10

This work studies the task of glossification, of which the aim is to em transcribe natural spoken language sentences for the Deaf (hard-of-hearing) community to ordered sign language glosses. Previous sequence-to-sequence language models trained with paired sentence-gloss data often fail to capture the rich connections between the two distinct languages, leading to unsatisfactory transcriptions. We observe that despite different grammars, glosses effectively simplify sentences for the ease of deaf communication, while sharing a large portion of vocabulary with sentences. This has motivated us to implement glossification by executing a collection of editing actions, e.g. word addition, deletion, and copying, called editing programs, on their natural spoken language counterparts. Specifically, we design a new neural agent that learns to synthesize and execute editing programs, conditioned on sentence contexts and partial editing results. The agent is trained to imitate minimal editing programs, while exploring more widely the program space via policy gradients to optimize sequence-wise transcription quality. Results show that our approach outperforms previous glossification models by a large margin.

poster session 2 @ red 6, poster session 5 @ red 6, poster session 2, poster session 5

Studies have shown that some Natural Language Processing (NLP) systems encode and replicate harmful biases with potential adverse ethical effects in our society. In this article, we propose an approach for identifying gender and racial stereotypes in word embeddings trained on judicial opinions from U.S. case law. Embeddings containing stereotype information may cause harm when used by downstream systems for classification, information extraction, question answering, or other machine learning systems used to build legal research tools. We first explain how previously proposed methods for identifying these biases are not well suited for use with word embeddings trained on legal opinion text. We then propose a domain adapted method for identifying gender and racial biases in the legal domain. Our analyses using these methods suggest that racial and gender biases are encoded into word embeddings trained on legal opinions. These biases are not mitigated by exclusion of historical data, and appear across multiple large topical areas of the law. Implications for downstream systems that use legal opinion word embeddings and suggestions for potential mitigation strategies based on our observations are also discussed.

poster session 2 @ red 6, poster session 5 @ red 6, oral session 2 @ red 6, poster session 2, poster session 5, oral session 2

In this paper, a new deep learning network named as graph-based point tracker (GPT) is proposed for 3D object tracking in point clouds. GPT is not based on Siamese network applied to template and search area, but it is based on the transfer of target clue from the template to the search area. GPT is end-to-end trainable. GPT has two new modules: graph feature augmentation (GFA) and improved target clue (ITC) module. The key idea of GFA is to exploit one-to-many relationship between template and search area points using a bipartite graph. In GFA, edge features of the bipartite graph are generated by transferring the target clues of template points to search area points through edge convolution. It captures the relationship between template and search area points effectively from the perspective of geometry and shape of two point clouds. The second module is ITC. The key idea of ITC is to embed the information of the center of the target into the edges of the bipartite graph via Hough voting, strengthening the discriminative power of GFA. Both modules significantly contribute to the improvement of GPT by transferring geometric and shape information including target center from target template to search area effectively. Experiments on the KITTI tracking dataset show that GPT achieves state-of-the-art performance and can run in real-time.

poster session 5 @ red 1, poster session 9 @ red 1, poster session 5, poster session 9

Modern recommender systems are required to adapt to the change in popularity of items and user preferences. Such problem is known as the temporal dynamics problem, and it is one of the main challenges in recommender system modeling. Different from the popular recurrent modeling approach, we propose a new solution named LeapRec to the temporal dynamic problem by using trajectory-based meta-learning to model time dependencies. LeapRec characterizes temporal dynamics by two complement components named global time leap (GTL) and ordered time leap (OTL). By design, GTL learns long-term patterns by finding the shortest learning path across unordered temporal data. Cooperatively, OTL learns short-term patterns by considering the sequential nature of the temporal data. Our experimental results show that LeapRec consistently outperforms the state-of-the-art methods on several datasets and recommendation metrics. Furthermore, we provide an empirical study of the interaction between GTL and OTL, showing the effects of long- and short-term modeling.

poster session 5 @ blue 5, poster session 9 @ blue 5, poster session 5, poster session 9

The existence of EFX allocations of goods is a major open problem in fair division, even for additive valuations. The current state of the art is that no setting where EFX allocations are impossible is known, and yet, existence results are known only for very restricted settings, such as: (i) agents with identical valuations, (ii) 2 agents, and (iii) 3 agents with additive valuations. It is also known that EFX exists if one can leave n-1 items unallocated, where n is the number of agents. We develop new techniques that allow us to push the boundaries of the enigmatic EFX problem beyond these known results, and (arguably) to simplify proofs of earlier results. Our main result is that every setting with 4 additive agents admits an EFX allocation that leaves at most a single item unallocated. Beyond our main result, we introduce a new class of valuations, termed nice cancelable, which includes additive, unit-demand, budget-additive and multiplicative valuations, among others. Using our new techniques, we show that both our results and previous results for additive valuations extend to nice cancelable valuations.

poster session 5 @ blue 6, poster session 8 @ blue 6, oral session 8 @ blue 6, poster session 5, poster session 8, oral session 8

Standard contrastive learning approaches usually require a large number of negatives for effective unsupervised learning and often exhibit slow convergence. We suspect this behavior is due to the suboptimal selection of negatives used for offering contrast to the positives. We counter this difficulty by taking inspiration from support vector machines (SVMs) to present max-margin contrastive learning (MMCL). Our approach selects negatives as the sparse support vectors obtained via a quadratic optimization problem, and contrastiveness is enforced by maximizing the decision margin. As SVM optimization can be computationally demanding, especially in an end-to-end setting, we present simplifications that alleviate the computational burden. We validate our approach on standard vision benchmark datasets, demonstrating better performance in unsupervised representation learning over state-of-the-art, while having better empirical convergence properties.

poster session 5 @ red 4, poster session 10 @ red 4, poster session 5, poster session 10

With ever-increasing dataset sizes, subset selection techniques are becoming increasingly important for a plethora of tasks. It is often necessary to guide the subset selection to achieve certain desiderata, which includes focusing or targeting certain data points, while avoiding others. Examples of such problems include: i)targeted learning, where the goal is to find subsets with rare classes or rare attributes on which the model is under performing, and ii)guided summarization, where data (e.g.,image collection, text, document or video) is summarized for quicker human consumption with specific additional user in-tent. Motivated by such applications, we present PRISM, a rich class of PaRameterIzed Submodular information Measures. Through novel functions and their parameterizations, PRISM offers a variety of modeling capabilities that enable a trade-off between desired qualities of a subset like diversity or representation and similarity/dissimilarity with a set of data points. We demonstrate how PRISM can be applied to the two real-world problems mentioned above, which require guided subset selection. In doing so, we show that PRISM interestingly generalizes some past work, therein reinforcing its broad utility. Through extensive experiments on diverse datasets, we demonstrate the superiority of PRISM over the state-of-the-art in targeted learning and in guided image-collection summarization.

poster session 5 @ blue 2, poster session 10 @ blue 2, poster session 5, poster session 10

Few-shot action recognition aims to recognize novel action classes (query) using just a few samples (support). The majority of current approaches follow the metric learning paradigm, which learns to compare the similarity between videos. Recently, it has been observed that directly measuring this similarity is not ideal since different action instances may show distinctive temporal distribution, resulting in severe misalignment issues across query and support videos. In this paper, we arrest this problem from two distinct aspects -- action duration misalignment and action evolution misalignment. We address them sequentially through a Two-stage Action Alignment Network (TA2N). The first stage locates the action by learning a temporal affine transform, which warps each video feature to its action duration while dismissing the action-irrelevant feature (e.g. background). Next, the second stage coordinates query feature to match the spatial-temporal action evolution of support by performing temporally rearrange and spatially offset prediction. Extensive experiments on benchmark datasets show the potential of the proposed method in achieving state-of-the-art performance for few-shot action recognition.

poster session 5 @ red 1, poster session 9 @ red 1, poster session 5, poster session 9

Single image deraining is an important and challenging task for some downstream artificial intelligence applications such as video surveillance and self-driving systems. Most of the existing deep-learning-based methods constrain the network to generate derained images but few of them explore features from intermediate layers, different levels, and different modules which are beneficial for rain streaks removal. In this paper, we propose a high-order collaborative network with multi-scale compact constraints and a bidirectional scale-content similarity mining module to exploit features from deep networks externally and internally for rain streaks removal. Externally, we design a deraining framework with three sub-networks trained in a collaborative manner, where the bottom network transmits intermediate features to the middle network which also receives shallower rainy features from the top network and sends features back to the bottom network. Internally, we enforce multi-scale compact constraints on the intermediate layers of deep networks to learn useful features via a Laplacian pyramid. Further, we develop a bidirectional scale-content similarity mining module to explore features at different scales in a down-to-up and up-to-down manner. To improve the model performance on real-world images, we propose an online-update learning approach, which uses real-world rainy images to fine-tune the network and update the deraining results in a self-supervised manner. Extensive experiments demonstrate that our proposed method performs favorably against eleven state-of-the-art methods on five public synthetic datasets and one real-world dataset. The source code can be found at: \url{https://supercong94.wixsite.com/supercong94}.

poster session 5 @ red 2, poster session 12 @ red 2, poster session 5, poster session 12

Neural language models show vulnerability to adversarial examples which are semantically similar to their original counterparts with a few words replaced by their synonyms. A common way to improve model robustness is adversarial training which follows two steps—collecting adversarial examples by attacking a target model, and fine-tuning the model on the augmented dataset with these adversarial examples. The objective of traditional adversarial training is to make a model produce the same correct predictions on an original/adversarial example pair. However, the consistency between model decision-makings on two similar texts is ignored. We argue that a robust model should behave consistently on original/adversarial example pairs, that is making the same predictions (what) based on the same reasons (how) which can be reflected by consistent interpretations. In this work, we propose a novel feature-level adversarial training method named FLAT. FLAT aims at improving model robustness in terms of both predictions and interpretations. FLAT incorporates variational word masks in neural networks to learn global word importance and play as a bottleneck teaching the model to make predictions based on important words. FLAT explicitly shoots at the vulnerability problem caused by the mismatch between model understandings on the replaced words and their synonyms in original/adversarial example pairs by regularizing the corresponding global word importance scores. Experiments show the effectiveness of FLAT in improving the robustness with respect to both predictions and interpretations of four neural network models (LSTM, CNN, BERT, and DeBERTa) to two adversarial attacks on four text classification tasks. The models trained via FLAT also show better robustness than baseline models on unforeseen adversarial examples across different attacks.

poster session 5 @ red 5, poster session 12 @ red 5, oral session 5 @ red 5, poster session 5, poster session 12, oral session 5

In this work, we focus on a very practical problem: image segmentation under rain conditions. Image deraining is a classic low-level restoration task, while image segmentation is a typical high-level understanding task. Most of the existing methods intuitively employ the bottom-up paradigm by taking deraining as a preprocessing step for subsequent segmentation. However, our statistical analysis indicates that not only deraining would benefit segmentation (bottom-up), but also segmentation would further improve deraining performance (top-down) in turn. This motivates us to solve the rainy image segmentation task within a novel top-down and bottom-up unified paradigm, in which two sub-tasks are alternatively performed and collaborated with each other. Specifically, the bottom-up procedure yields both clearer images and rain-robust features from both image and feature domains, so as to ease the segmentation ambiguity caused by rain streaks. The top-down procedure adopts semantics to adaptively guide the restoration for different contents via a novel multi-path semantic attentive module (SAM). Thus the deraining and segmentation could boost the performance of each other cooperatively and progressively. Extensive experiments and ablations demonstrate that the proposed method outperforms the state-of-the-art on rainy image segmentation.

poster session 5 @ red 2, poster session 12 @ red 2, poster session 5, poster session 12

A 6-hour early detection of sepsis leads to a significant increase in the chance of surviving it. Previous sepsis early detection studies have focused on improving the performance of supervised learning algorithms while ignoring the potential correlation in data mining, and there was no reliable method to deal with the problem of incomplete data. In this paper, we proposed the Denoising Transformer AutoEncoder (DTAE) for the first time combining transformer and unsupervised learning. DTAE can learn the correlation of the features required for early detection of sepsis without the label. This method can effectively solve the problems of data sparsity and noise and discover the potential correlation of features by adding DTAE enhancement module without modifying the existing algorithms. Finally, the experimental results show that the proposed method improves the existing algorithms and achieves the best results of early detection.

poster session 5 @ red 2, poster session 12 @ red 2, poster session 5, poster session 12

Graph representation learning received increasing attentions in recent years. Most of the existing methods ignore the complexity of the graph structures and restrict graphs in a single constant-curvature representation space, which is only suitable to particular kinds of graph structure indeed. Additionally, these methods follow the supervised or semi-supervised learning paradigm, and thereby notably limit their deployment on the unlabeled graphs in real applications. To address these aforementioned limitations, we take the first attempt to study the self-supervised graph representation learning in the mixed-curvature spaces. In this paper, we present a novel Self-supervised Mixed-curvature Graph Neural Network (SelfMGNN). Instead of working on one single constant-curvature space, we construct a mixed-curvature space via the Cartesian product of multiple Riemannian component spaces and design hierarchical attention mechanisms for learning and fusing the representations across these component spaces. To enable the self-supervisd learning, we propose a novel dual contrastive approach. The constructed mixed-curvature space actually provides multiple Riemannian views for the contrastive learning. We introduce a Riemannian projector to reveal these views, and utilize a well-designed Riemannian discriminator for the single-view and cross-view contrastive learning within and across the Riemannian views. Finally, extensive experiments show that SelfMGNN captures the complicated graph structures in reality and outperforms state-of-the-art baselines.

poster session 5 @ blue 5, poster session 9 @ blue 5, oral session 9 @ blue 5, poster session 5, poster session 9, oral session 9

Fairness related to locations (i.e., "where") is critical for the use of machine learning in a variety of societal domains involving spatial datasets (e.g., agriculture, disaster response, urban planning). Spatial biases incurred by learning, if left unattended, may cause or exacerbate unfair distribution of resources, social division, spatial disparity, etc. The goal of this work is to develop statistically-robust formulations and model-agnostic learning strategies to understand and promote spatial fairness. The problem is challenging as locations are often from continuous spaces with no well-defined categories (e.g., gender), and statistical conclusions from spatial data are fragile to changes in spatial partitioning and scales. Existing studies in fairness-driven learning have generated valuable insights related to non-spatial factors including race, gender, education level, etc., but research to mitigate location related biases still remain in its infancy, leaving the main challenges unaddressed. To bridge the gap, we first propose a robust space-as-distribution (SPAD) representation of spatial fairness to reduce statistical sensitivity related to partitioning and scales in continuous space. Furthermore, we propose a new SPAD-based stochastic strategy to efficiently optimize over an extensive distribution of fairness criteria, and a bi-level training framework to enforce fairness via adaptive adjustment of priorities among locations. Experiments and case studies on real-world agricultural monitoring show that SPAD can effectively reduce sensitivity in spatial fairness evaluation and the proposed stochastic bi-level training framework can greatly improve fairness for different base learning models.

poster session 2 @ red 6, poster session 5 @ red 6, poster session 2, poster session 5

Many real-world scenarios, such as human activity recognition (HAR) in IoT, can be formalized as a multi-task multi-view learning problem. Each specific task consists of multiple shared feature views collected from multiple sources, either homogeneous or heterogeneous. Common among recent approaches is to employ a typical hard/soft sharing strategy at the initial phase separately for each view across tasks to uncover common knowledge, underlying the assumption that all views are conditionally independent. On the one hand, multiple views across tasks possibly relate to each other under practical situations. On the other hand, supervised methods might be insufficient when labeled data is scarce. To tackle these challenges, we introduce a novel framework ASM2TV for semi-supervised multi-task multi-view learning. We present a new perspective named gating control policy, a learnable task-view-interacted sharing policy that adaptively selects the most desirable candidate shared block for any view across any task, which uncovers more fine-grained task-view-interacted relatedness and improves inference efficiency. Significantly, our proposed gathering consistency adaption procedure takes full advantage of large amounts of unlabeled fragmented time-series, making it a general framework that accommodates a wide range of applications. Experiments on two diverse real-world HAR benchmark datasets collected from various subjects and sources demonstrate our framework's superiority over other state-of-the-arts. Anonymous codes are available at https://github.com/zachstarkk/ASM2TV.

poster session 5 @ blue 3, poster session 10 @ blue 3, poster session 5, poster session 10

We consider the problem of automatically constructing computer programs from input-output examples. We investigate how to augment probabilistic and neural program synthesis methods with new search algorithms, proposing a framework called distribution-based search. Within this framework, we introduce two new search algorithms: Heap Search, an enumerative method, and SQRT Sampling, a probabilistic method. We prove certain optimality guarantees for both methods, show how they integrate with probabilistic and neural techniques, and demonstrate how they can operate at scale across parallel compute environments. Collectively these findings offer theoretical and applied studies of search algorithms for program synthesis that integrate with recent developments in machine-learned program synthesizers.

poster session 5 @ blue 2, poster session 10 @ blue 2, oral session 5 @ blue 2, poster session 5, poster session 10, oral session 5

Deep semi-supervised learning (SSL) aims to utilize a sizeable unlabeled set to train deep networks, thereby reducing the dependence on labeled instances. However, the unlabeled set often carries unseen classes that cause the deep SSL algorithm to lose generalization. Previous works focus on the data level that they attempt to remove unseen class data or assign lower weight to them but could not eliminate their adverse effects on the SSL algorithm. Rather than focusing on the data level, this paper turns attention to the model parameter level. We find that only partial parameters are essential for seen-class classification, termed safe parameters. In contrast, the other parameters tend to fit irrelevant data, termed harmful parameters. Driven by this insight, we propose Safe Parameter Learning (SPL) to discover safe parameters and make the harmful parameters inactive, such that we can mitigate the adverse effects caused by unseen-class data. Specifically, we firstly design an effective strategy to divide all parameters in the pre-trained SSL model into safe and harmful ones. Then, we introduce a bi-level optimization strategy to update the safe parameters and kill the harmful parameters. Extensive experiments show that SPL outperforms the state-of-the-art SSL methods on all the benchmarks by a large margin. Moreover, experiments demonstrate that SPL can be integrated into the most popular deep SSL networks and be easily extended to handle other cases of class distribution mismatch.

poster session 5 @ red 4, poster session 10 @ red 4, oral session 5 @ red 4, poster session 5, poster session 10, oral session 5

In this paper, we propose a new joint object detection and tracking (JoDT) framework for 3D object detection and tracking based on camera and LiDAR sensors. The proposed method, referred to as 3D DetecTrack, enables the detector and tracker to cooperate to generate a spatio-temporal representation of the camera and LiDAR data, with which 3D object detection and tracking are then performed. The detector constructs the spatio-temporal features via the weighted temporal aggregation of the spatial features obtained by the camera and LiDAR fusion. Then, the detector reconfigures the initial detection results using information from the tracklets maintained up to the previous time step. Based on the spatio-temporal features generated by the detector, the tracker associates the detected objects with previously tracked objects using a graph neural network (GNN). We devise a fully-connected GNN facilitated by a combination of rule-based edge pruning and attention-based edge gating, which exploits both spatial and temporal object contexts to improve tracking performance. The experiments conducted on both KITTI and nuScenes benchmarks demonstrate that the proposed 3D DetecTrack achieves significant improvements in both detection and tracking performances over baseline methods and achieves state-of-the-art performance among existing methods through collaboration between the detector and tracker.

poster session 5 @ red 1, poster session 9 @ red 1, poster session 5, poster session 9

Neuroimaging studies have identified multiple brain regions that are associated with semantic and syntactic processing when comprehending language. However, existing methods cannot explore the neural correlates of fine-grained word syntactic features, such as part-of-speech and dependency relations. This paper proposes an alternative framework to study how different word syntactic features are represented in the brain. To separate each syntactic feature, we propose a feature elimination method, called Mean Vector Null space Projection (MVNP). This method can remove a specific feature from word representations, resulting in one-feature-removed representations. Then we respectively associate one-feature-removed and the original word vectors with brain imaging data to explore how the brain represents the removed feature. This paper for the first time studies the cortical representations of multiple fine-grained syntactic features simultaneously and suggests some possible contributions of several brain regions to the complex division of syntactic processing. These findings indicate that the brain foundations of syntactic information processing might be broader than those suggested by classical studies.

poster session 5 @ red 5, poster session 12 @ red 5, poster session 5, poster session 12

Planar object tracking plays an important role in AI applications, such as robotics, visual servoing, and visual SLAM. Although the previous planar trackers work well in most scenarios, it is still a challenging task due to the rapid motion and large transformation between two consecutive frames. The essential reason behind this problem is that the condition number of such a non-linear system changes unstably when the searching range of the homography parameter space becomes larger. To this end, we propose a novel Homography Decomposition Networks (HDN) approach that drastically reduces and stabilizes the condition number by decomposing the homography transformation into two groups. Specifically, a similarity transformation estimator is designed to predict the first group robustly by a deep convolution equivariant network. By taking advantage of the scale and rotation estimation with high confidence, a residual transformation is estimated by a simple regression model. Furthermore, the proposed end-to-end network is trained in a semi-supervised fashion. Extensive experiments show that our proposed approach outperforms the state-of-the-art planar tracking methods at a large margin on the challenging POT, UCSB and POIC datasets.

poster session 5 @ red 1, poster session 9 @ red 1, oral session 9 @ red 1, poster session 5, poster session 9, oral session 9

Multiple Instance Learning (MIL) is a weakly supervised learning problem where the aim is to assign labels to sets or bags of instances, as opposed to traditional supervised learning where each instance is assumed to be {\em independent and identically distributed} (IID) and is to be labeled individually. Recent work has shown promising results for neural network models in the MIL setting. Instead of focusing on each instance, these models are trained in an end-to-end fashion to learn effective bag-level representations by suitably combining permutation invariant pooling techniques with neural architectures. In this paper, we consider modelling the interactions between bags using a graph and employ Graph Neural Networks (GNNs) to facilitate end-to-end learning. Since a meaningful graph representing dependencies between bags is rarely available, we propose to use a Bayesian GNN framework that can generate a likely graph structure for scenarios where there is uncertainty in the graph or when no graph is available. Empirical results demonstrate the efficacy of the proposed technique for several MIL benchmark tasks and a distribution regression task.

poster session 5 @ red 4, poster session 10 @ red 4, poster session 5, poster session 10

The vast amount of data captured by robots in large-scale environments brings the computing and storage bottlenecks to the typical methods of modeling the spaces the robots travel in. In order to efficiently construct a compact terrain model from uncertain, incomplete point cloud data of large-scale environments, in this paper, we first propose a novel feature mapping method, named random mapping, based on the fast random construction of base functions, which can efficiently project the messy points in the low-dimensional space into the high-dimensional space where the points are approximately linearly distributed. Then, in this mapped space, we propose to learn a continuous linear regression model to represent the terrain. We show that this method can model the environments in much less computation time, memory consumption and access time with high accuracy. Furthermore, the models possess the generalization capabilities comparable to the performances on the training set, and its inference accuracy gradually increases as the random mapping dimension increases. To better solve the large-scale environmental modeling problem, we adopt the idea of parallel computing to train the models. This strategy greatly reduce the wall-clock time of calculation without losing much accuracy. Experiments show the effectiveness of random mapping method and effects of some important parameters on its performances. Moreover, we evaluate the proposed terrain modeling method based on the random mapping method and compare its performances with popular typical methods and state-of-art methods.

poster session 5 @ red 3, poster session 12 @ red 3, poster session 5, poster session 12

Human-Object Interaction (HOI) detection is a fundamental task in high-level human-centric scene understanding. We propose PhraseHOI, containing a HOI branch and a novel phrase branch, to leverage language prior and improve relation expression. Specifically, the phrase branch is supervised by semantic embeddings, whose ground truths are automatically converted from the original HOI annotations without extra human efforts. Meanwhile, a novel label composition method is proposed to deal with the long-tailed problem in HOI, which composites novel phrase labels by semantic neighbors. Further, to optimize the phrase branch, a loss composed of a distilling loss and a balanced triplet loss is proposed. Extensive experiments are conducted to prove the effectiveness of the proposed PhraseHOI, which achieves significant improvement over the baseline and surpasses previous state-of-the-art methods on Full and NonRare on the challenging HICO-DET benchmark.

poster session 5 @ red 3, poster session 12 @ red 3, poster session 5, poster session 12

Vision transformers have recently demonstrated great success in various computer vision tasks, motivating a tremendously increased interest in their deployment into many real-world IoT applications. However, powerful ViTs are often too computationally expensive to be fitted onto real-world resource-constrained platforms, due to (1) their quadratically increased complexity with the number of input tokens and (2) their overparameterized self-attention heads and model depth. In parallel, different images are of varied complexity and their different regions can contain various levels of visual information, e.g., a sky background is not as informative as a foreground object in object classification tasks, indicating that treating those regions equally in terms of model complexity is unnecessary while such opportunities for trimming down ViTs' complexity have not been fully exploited. To this end, we propose a Multi-grained Input-Adaptive Vision Transformer framework dubbed MIA-Former that can input-adaptively adjust the structure of ViTs at three coarse-to-fine-grained granularities (i.e., model depth and the number of model heads/tokens). In particular, our MIA-Former adopts a low-cost network trained with a hybrid supervised and reinforcement learning method to skip the unnecessary layers, heads, and tokens in an input adaptive manner, reducing the overall computational cost. Furthermore, an interesting side effect of our MIA-Former is that its resulting ViTs are naturally equipped with improved robustness against adversarial attacks over their static counterparts, because MIA-Former's multi-grained dynamic control improves the model diversity similar to the effect of ensemble and thus increases the difficulty of adversarial attacks against all its sub-models. Extensive experiments and ablation studies validate that the proposed MIA-Former framework can (1) effectively allocate adaptive computation budgets to the difficulty of input images, achieving state-of-the-art (SOTA) accuracy-efficiency trade-offs, e.g., up to 16.5\% computation savings with the same or even a higher accuracy compared with the SOTA dynamic transformer models, and (2) boost ViTs' robustness accuracy under various adversarial attacks over their vanilla counterparts by 2.4\% and 3.0\%, respectively.

poster session 5 @ blue 3, poster session 10 @ blue 3, poster session 5, poster session 10

The advancement in Image-to-Image translation techniques using generative Deep Learning-based approaches has shown promising results for the challenging task of inpainting-based 3D view synthesis. At the same time, even the current 3D view synthesis methods often create distorted structures or blurry textures inconsistent with surrounding areas. We analyzed the recently proposed algorithms for inpainting-based 3D view synthesis and observed that these algorithms no longer produce stretching and black holes. However, the existing databases such as IETR, IRCCyN, and IVY have 3D-generated views with these artifacts. This observation suggests that the existing 3D view synthesis quality assessment algorithms can not judge the quality of most recent 3D synthesized views. With this view, through this abstract, we analyze the need for a new large-scale database and a new perceptual quality metric oriented for 3D views using a test dataset.

poster session 5 @ red 1, poster session 9 @ red 1, poster session 5, poster session 9

When aggregating logically interconnected judgments from n agents, the result might be logically inconsistent. This phenomenon is known as the doctrinal paradox, which plays a central role in the field of judgment aggregation. Previous work has mostly focused on the worst-case analysis of the doctrinal paradox, which has lead to many impossibility results. Little is known about its practical occurrence, except for the study on its likelihood under a few i.i.d. distributions. In this paper, we characterize the likelihood of the doctrinal paradox under a much more general and realistic model called smoothed social choice framework introduced in a NeurIPS 2020 paper. In the framework, agents' ground truth judgments can be arbitrarily correlated, while the noises are independent. Our main theorem states that under mild conditions, the smoothed likelihood of the doctrinal paradox is either $0$, $\exp(-\Theta(n))$, $\Theta(n^{-1/2})$ or $\Theta(1)$. This not only answers open questions by List in 2005, but also draws clear lines between situations with frequent and with vanishing paradoxes.

poster session 5 @ blue 6, poster session 8 @ blue 6, poster session 5, poster session 8

In AI-assisted decision-making, effective hybrid (human-AI) teamwork is not solely dependent on AI performance alone, but also on its impact on human decision-making. While prior work studies the effects of model accuracy on humans, we endeavour here to investigate the complex dynamics of how both a model's predictive performance and bias may transfer to humans in a recommendation-aided decision task. We consider the domain of ML-assisted hiring, where humans---operating in a constrained selection setting---can choose whether they wish to utilize a trained model's inferences to help select candidates from written biographies. We conduct a large-scale user study leveraging a re-created dataset of real bios from prior work, where humans predict the ground truth occupation of given candidates with and without the help of three different NLP classifiers (random, bag-of-words, and deep neural network). Our results demonstrate that while high-performance models significantly improve human performance in a hybrid setting, some models mitigate hybrid bias while others accentuate it. We examine these findings through the lens of decision conformity and observe that our model architecture choices have an impact on human-AI conformity and bias, motivating the explicit need to assess these complex dynamics prior to deployment.

poster session 2 @ red 6, poster session 5 @ red 6, oral session 2 @ red 6, poster session 2, poster session 5, oral session 2

Animals are equipped with a rich innate repertoire of sensory, behavioral and motor skills, which allows them to interact with the world immediately after birth. At the same time, many behaviors are highly adaptive and can be tailored to specific environments by means of learning. In this work, we use mathematical analysis and the framework of memory-based meta-learning (or 'learning to learn') to answer when it is beneficial to learn such an adaptive strategy and when to hard-code a heuristic behavior. We find that the interplay of ecological uncertainty, task complexity and the agents' lifetime has crucial effects on the meta-learned amortized Bayesian inference performed by an agent. There exist two regimes: One in which meta-learning yields a learning algorithm that implements task-dependent information-integration and a second regime in which meta-learning imprints a heuristic or 'hard-coded' behavior. Further analysis reveals that non-adaptive behaviors are not only optimal for aspects of the environment that are stable across individuals, but also in situations where an adaptation to the environment would in fact be highly beneficial, but could not be done quickly enough to be exploited within the remaining lifetime. Hard-coded behaviors should hence not only be those that always work, but also those that are too complex to be learned within a reasonable time frame.

poster session 5 @ red 4, poster session 10 @ red 4, poster session 5, poster session 10

Federated learning (FL) is a training technique that enables client devices to jointly learn a shared model by aggregating locally-computed models without exposing their raw data. While most of the existing work focuses on improving the FL model accuracy, in this paper, we focus on the improving the training efficiency, which is often a hurdle for adopting FL in real-world applications. Specifically, we design an efficient FL framework which jointly optimizes model accuracy, processing latency and communication efficiency, all of which are primary design considerations for real implementation of FL. Inspired by the recent success of Multi-Agent Reinforcement Learning (MARL) in solving complex control problems, we present \textit{FedMarl}, an MARL-based FL framework which performs efficient run-time client selection. Experiments show that FedMarl can significantly improve model accuracy with much lower processing latency and communication cost.

poster session 5 @ blue 3, poster session 10 @ blue 3, poster session 5, poster session 10

Unsupervised learning is often used to uncover clusters in data. However, different kinds of noise may impede the discovery of useful patterns from real-world time-series data. In this work, we focus on mitigating the interference of interval censoring in the task of clustering for disease phenotyping. We develop a deep generative, continuous-time model of time-series data that clusters time-series while correcting for censorship time. We provide conditions under which clusters and the amount of delayed entry may be identified from data under a noiseless model. On synthetic data, we demonstrate accurate, stable, and interpretable results that outperform several benchmarks. On real-world clinical datasets of heart failure and Parkinson's disease patients, we study how interval censoring can adversely affect the task of disease phenotyping. Our model corrects for this source of error and recovers known clinical subtypes.

poster session 5 @ blue 4, poster session 7 @ blue 4, poster session 5, poster session 7

Generally, humans are more skilled at perceiving differences between high-quality (HQ) and low-quality (LQ) images than directly judging the quality of a single LQ image. This situation also applies to image quality assessment (IQA). Although recent no-reference (NR-IQA) methods have made great progress to predict image quality free from the reference image, they still have the potential to achieve better performance since HQ image information is not fully exploited. In contrast, full-reference (FR-IQA) methods tend to provide more reliable quality evaluation, but its practicability is affected by the requirement for pixel-level aligned reference images. To address this, we firstly propose the content-variant reference method via knowledge distillation (CVRKD-IQA). Specifically, we use non-aligned reference (NAR) images to introduce various prior distributions of high-quality images. The comparisons of distribution differences between HQ and LQ images can help our model better assess the image quality. Further, the knowledge distillation transfers more HQ-LQ distribution difference information from the FR-teacher to the NAR-student and stabilizing CVRKD-IQA performance. Moreover, to fully mine the local-global combined information, while achieving faster inference speed, our model directly processes multiple image patches from the input with the MLP-mixer. Cross-dataset experiments verify that our model can outperform all NAR/NR-IQA SOTAs, even reach comparable performance than FR-IQA methods on some occasions. Since the content-variant and non-aligned reference HQ images are easy to obtain, our model can support more IQA applications with its robustness to content variations. Our code is available: https://github.com/guanghaoyin/CVRKD-IQA.

poster session 5 @ red 2, poster session 12 @ red 2, oral session 12 @ red 2, poster session 5, poster session 12, oral session 12

Student modeling sits at the epicenter of adaptive learning technology. In contrast to the voluminous work on student modeling for well-defined domains such as algebra, there has been little research on student modeling in programming (SMP). As part of developing adaptive learning environments for programming, we are focusing on two essential SMP tasks: program classification and early prediction of student success. Due to the unbounded solution spaces for open-ended programming exercises, researchers often face the issue of data scarcity in SMP. In this work, we propose a Cross-Lingual Adversarial Domain Adaptation (CrossLing) framework that can leverage a large programming dataset to learn features that can improve SMP's build using a much smaller dataset in a different programming language. Our framework maintains one globally invariant latent representation across both datasets via an adversarial learning process, as well as allocating domain-specific models for each dataset to extract local latent representations that cannot and should not be united. By separating globally-shared representations from domain-specific representations, our framework outperforms existing state-of-the-art methods for both SMP tasks.

poster session 5 @ blue 4, poster session 7 @ blue 4, oral session 5 @ blue 4, poster session 5, poster session 7, oral session 5

Deep Convolutional Neural Networks (CNNs)-based methods have achieved significant breakthroughs in the task of single image shadow removal. However, the performance of these methods remains limited for several reasons. First, the existing shadow illumination model ignores the spatially-variant property of the shadow images, hindering their further performance. Second, most deep CNNs-based methods directly estimate the shadow-free results from the input shadow images like a black box, thus losing the desired interpretability. To address these issues, we first propose a new shadow illumination model for the shadow removal task. This new shadow illumination model ensures the identity mapping among non-shadow regions, and adaptively performs fine-grained spatial mapping between shadow regions and their references. Then, based on the shadow illumination model, we reformulate the shadow removal task as a variational optimization problem. To effectively solve the variational problem, we design an iterative algorithm and unfold it into a deep network, naturally increasing the interpretability of the deep model. Experiments show that our method could achieve state-of-the-art performance with 46% parameters, 21% floating point of operations (FLOPs) and over 17 times faster than SOTA method DHAN \cite{cun2020towards}.

poster session 5 @ red 2, poster session 12 @ red 2, poster session 5, poster session 12

Spiking neural networks (SNNs) with leaky integrate and fire (LIF) neurons, can be operated in an event-driven manner and have internal states to retain information over time, providing opportunities for energy-efficient neuromorphic computing, especially on edge devices. Note, however, many representative works on SNNs do not fully demonstrate the usefulness of their inherent recurrence (membrane potential retaining information about the past) for sequential learning. Most of the works train SNNs to recognize static images by artificially expanded input representation in time through rate coding. We show that SNNs can be trained for practical sequential tasks and propose modifications to a network of LIF neurons that enable internal states to learn long sequences and make their inherent recurrence resilient to the vanishing gradient problem. We then develop a training scheme to train the proposed SNNs with improved inherent recurrence dynamics. Our training scheme allows spiking neurons to produce multi-bit outputs (as opposed to binary spikes) which help mitigate the mismatch between a derivative of spiking neurons' activation function and a surrogate derivative used to overcome spiking neurons' non-differentiability. Our experimental results indicate that the proposed SNN architecture on TIMIT and LibriSpeech 100h speech recognition dataset yields accuracy comparable to that of LSTMs (within 1.10% and 0.36%, respectively), but with 2x fewer parameters than LSTMs. The sparse SNN outputs also lead to 10.13x and 11.14x savings in multiplication operations compared to GRUs, which is generally considered as a lightweight alternative to LSTMs, on TIMIT and LibriSpeech 100h datasets, respectively.

poster session 5 @ blue 4, poster session 7 @ blue 4, oral session 5 @ blue 4, poster session 5, poster session 7, oral session 5

The success of the state-of-the-art video deblurring methods stems mainly from implicit or explicit estimation of alignment among the adjacent frames for latent video restoration. However, due to the influence of the blur effect, estimating the alignment information from the blurry adjacent frames is not a trivial task. Inaccurate estimations will interfere the following frame restoration. Instead of estimating alignment information, we propose a simple and effective deep Recurrent Neural Network with Multi-scale Bi-directional Propagation (RNN-MBP) to effectively propagate and gather the information from unaligned neighboring frames for better video deblurring. Specifically, we build a Multi-scale Bi-directional Propagation~(MBP) module with two U-Net RNN cells which can directly exploit the inter-frame information from unaligned neighboring hidden states by integrating them in different scales. Moreover, to better evaluate the proposed algorithm and existing state-of-the-art methods on real-world blurry scenes, we also create a Real-World Blurry Video Dataset (RBVD) by a well-designed Digital Video Acquisition System (DVAS) and use it as the training and evaluation dataset. Extensive experimental results demonstrate that the proposed RBVD dataset effectively improve the performance of existing algorithms on real-world blurry videos, and the proposed algorithm performs favorably against the state-of-the-art methods on three typical benchmarks.

poster session 5 @ red 2, poster session 12 @ red 2, poster session 5, poster session 12

Although many graph-based clustering methods attempt to model the stationary diffusion state in their objectives, their performance limits to using a predefined graph. We argue that the estimation of the stationary diffusion state can be achieved by gradient descent over neural networks. We specifically design the Stationary Diffusion State Neural Estimation (SDSNE) to exploit multiview structural graph information for co-supervised learning. We explore how to design a graph neural network specially for unsupervised multiview learning and integrate multiple graphs into an unified consensus graph by a shared self-attentional module. The view-shared self-attentional module utilizes the graph structure to learn a view-consistent global graph. Meanwhile, instead of using auto-encoder in most unsupervised learning graph neural networks, SDSNE uses a co-supervised strategy with structure information to supervise the model learning. The co-supervised strategy as the loss function guides SDSNE in achieving the stationary state. With the help of the loss and the self-attentional module, we learn to obtain a graph in which nodes in each connected component fully connect by the same weight. Experiments on several multiview datasets demonstrate effectiveness of SDSNE in terms of six clustering evaluation metrics.

poster session 5 @ red 4, poster session 10 @ red 4, oral session 5 @ red 4, poster session 5, poster session 10, oral session 5

Temporal grounding aims to localize a video moment which is semantically aligned with a given natural language query. Existing methods typically apply a detection or regression pipeline on the fused representation with the research focus on designing complicated prediction heads or fusion strategies. Instead, from a perspective on temporal grounding as a metric-learning problem, we present a Mutual Matching Network (MMN), to directly model the similarity between language queries and video moments in a joint embedding space. This new metric-learning framework enables fully exploiting negative samples from two new aspects: constructing negative cross-modal pairs in a mutual matching scheme and mining negative pairs across different videos. These new negative samples could enhance the joint representation learning of two modalities via cross-modal mutual matching to maximize their mutual information. Experiments show that our MMN achieves highly competitive performance compared with the state-of-the-art methods on four video grounding benchmarks. Based on MMN, we present a winner solution for the HC-STVG challenge of the 3rd PIC workshop. This suggests that metric learning is still a promising method for temporal grounding via capturing the essential cross-modal correlation in a joint embedding space. Code is available at https://github.com/MCG-NJU/MMN.

poster session 5 @ red 1, poster session 9 @ red 1, poster session 5, poster session 9

Recently, the scheme of model-X knockoffs was proposed as a promising solution to address controlled feature selection under high-dimensional finite-sample settings. However, the procedure of model-X knockoffs depends heavily on the coefficient-based feature importance and only concerns the control of false discovery rate (FDR). To further improve its adaptivity and flexibility, in this paper, we propose an error-based knockoff inference method by integrating the knockoff features, the error-based feature importance statistics, and the stepdown procedure together. The proposed inference procedure does not require specifying a regression model and can handle feature selection with theoretical guarantees on controlling false discovery proportion (FDP), FDR, or k-familywise error rate (k-FWER). Empirical evaluations demonstrate the competitive performance of our approach on both simulated and real data.

poster session 5 @ blue 1, poster session 10 @ blue 1, poster session 5, poster session 10

We study the problem of safe online convex optimization, where the action at each time step must satisfy a set of linear safety constraints. The goal is to select a sequence of actions to minimize the regret without violating the safety constraints at any time step (with high probability). The parameters that specify the linear safety constraints are unknown to the algorithm. The algorithm has access to only the noisy observations of constraints for the chosen actions. We propose an algorithm, called the Safe Online Projected Gradient Descent (SO-PGD) algorithm, to address this problem. We show that, under the assumption of the availability of a safe baseline action, the SO-PGD algorithm achieves a regret $O(T^{2/3})$. While there are many algorithms for online convex optimization (OCO) problems with safety constraints available in the literature, they allow constraint violations during learning/optimization, and the focus has been on characterizing the cumulative constraint violations. To the best of our knowledge, ours is the first work that provides an algorithm with provable guarantees on the regret, without violating the linear safety constraints (with high probability) at any time step.

poster session 5 @ blue 1, poster session 10 @ blue 1, poster session 5, poster session 10

With the abundance of available data, many enterprises seek to implement data-driven prescriptive analytics to help them make informed decisions. These prescriptive policies need to satisfy operational constraints, and proactively eliminate rule conflicts, both of which are ubiquitous in practice. It is also desirable for them to be simple and interpretable, so they can be easily verified and implemented. Existing approaches from the literature center around constructing variants of prescriptive decision trees to generate interpretable policies. However, none of the existing methods is able to handle constraints. In this paper, we propose a scalable method that solves the constrained prescriptive policy generation problem. We introduce a novel path-based mixed-integer program (MIP) formulation which identifies a (near) optimal policy efficiently via column generation. The policy generated can be represented as a multiway-split tree which is more interpretable and informative than binary-split trees due to its shorter rules. We demonstrate the efficacy of our method with extensive computational experiments on both synthetic and real datasets.

poster session 5 @ blue 2, poster session 10 @ blue 2, poster session 5, poster session 10

We study fairness through the lens of cooperative multi-agent learning. Our work is motivated by empirical evidence that naive maximization of team reward yields unfair outcomes for individual team members. To address fairness in multi-agent contexts, we introduce team fairness, a group-based fairness measure for multi-agent learning. We then prove that it is possible to enforce team fairness during policy optimization by transforming the team’s joint policy into an equivariant map. We refer to our multi-agent learning strategy as Fairness through Equivariance (Fair-E) and demonstrate its effectiveness empirically. We then introduce Fairness through Equivariance Regularization (Fair-ER) as a soft-constraint version of Fair-E and show that it reaches higher levels of utility than Fair-E and fairer outcomes than non-equivariant policies. Finally, we present novel findings regarding the fairness-utility trade-off in multi-agent settings; showing that the magnitude of the trade-off is dependent on agent skill.

poster session 5 @ blue 1, poster session 10 @ blue 1, poster session 5, poster session 10

Differential Privacy offers strong guarantees such as immutable privacy under any post-processing. In this work, we propose a differentially private mechanism called PrivateMail for performing supervised manifold learning. We then apply it to the use case of private image retrieval to obtain nearest matches to a client’s target image from a server’s database. PrivateMail releases the target image as part of a differentially private manifold embedding. We give bounds on the global sensitivity of the manifold learning map in order to obfuscate and release embeddings with differential privacy inducing noise. We show that PrivateMail obtains a substantially better performance in terms of the privacy-utility trade off in comparison to several baselines on various datasets. We share code for applying PrivateMail at http://tiny.cc/PrivateMail.

poster session 5 @ blue 2, poster session 10 @ blue 2, oral session 5 @ blue 2, poster session 5, poster session 10, oral session 5

We study the convergence rate of gradient-based local search methods for solving low-rank matrix recovery problems with general objectives in both symmetric and asymmetric cases, under the assumption of the restricted isometry property. First, we develop a new technique to verify the Polyak-Lojasiewicz inequality in a neighborhood of the global minimizers, which leads to a local linear convergence region for the gradient descent method. Second, based on the local convergence result and a sharp strict saddle property proven in this paper, we present two new conditions that guarantee the global linear convergence of the perturbed gradient descent method. The developed local and global convergence results provide much stronger theoretical guarantees than the existing results. As a by-product, this work significantly improves the existing bounds on the RIP constant required to guarantee the non-existence of spurious solutions.

poster session 5 @ blue 2, poster session 10 @ blue 2, poster session 5, poster session 10

Due to the surge of cloud-assisted AI services, the problem of designing resilient prediction serving systems that can effectively cope with stragglers/failures and minimize response delays has attracted much interest. The common approach for tackling this problem is replication which assigns the same prediction task to multiple workers. This approach, however, is very inefficient and incurs significant resource overheads. Hence, a learning-based approach known as parity model (ParM) has been recently proposed which learns models that can generate ``parities’’ for a group of predictions in order to reconstruct the predictions of the slow/failed workers. While this learning-based approach is more resource-efficient than replication, it is tailored to the specific model hosted by the cloud and is particularly suitable for a small number of queries (typically less than four) and tolerating very few (mostly one) number of stragglers. Moreover, ParM does not handle Byzantine adversarial workers. We propose a different approach, named Approximate Coded Inference (ApproxIFER), that does not require training of any parity models, hence it is agnostic to the model hosted by the cloud and can be readily applied to different data domains and model architectures. Compared with earlier works, ApproxIFER can handle a general number of stragglers and scales significantly better with the number of queries. Furthermore, ApproxIFER is robust against Byzantine workers. Our extensive experiments on a large number of datasets and model architectures also show significant degraded mode accuracy improvement by up to 58% over the parity model approaches.

poster session 5 @ blue 3, poster session 10 @ blue 3, poster session 5, poster session 10

We study the online influence maximization (OIM) problem in social networks, where in multiple rounds the learner repeatedly chooses seed nodes to generate cascades, observes the cascade feedback, and gradually learns the best seeds that generate the largest cascade. We focus on two major challenges in this paper. First, we work with \emph{node-level feedback} instead of \emph{edge-level feedback}. The edge-level feedback reveals all edges that pass through information in a cascade, where the node-level feedback only reveals the activated nodes with timestamps. The node-level feedback is arguably more realistic since in practice it is relatively easy to observe who is influenced but very difficult to observe from which relationship (edge) the influence comes from. Second, we use \emph{standard offline oracle} instead of \emph{offline pair-oracle}. To compute a good seed set for the next round, an offline pair-oracle finds the best seed set and the best parameters within the confidence region simultaneously, and such an oracle is difficult to compute due to the combinatorial core of OIM problem. So we focus on how to use the standard offline influence maximization oracle which finds the best seed set given the edge parameters as input. In this paper, we resolve these challenges for the two most popular diffusion models, the independent cascade (IC) and the linear threshold (LT) model. For the IC model, the past research only achieves edge-level feedback, while we present the first $\widetilde{O}(\sqrt{T})$-regret algorithm for the node-level feedback. For the first challenge above, we apply a novel adaptation of the maximum likelihood estimation (MLE) approach to learn the graph parameters and its confidence region (a confidence ellipsoid). For the second challenge, we adjust the update procedure to dissect the confidence ellipsoid into confidence intervals on each parameter, so that the standard offline influence maximization oracle is enough. For the LT model, a recent study only provides an OIM solution that meets the first challenge but still requires a pair-oracle. In this paper, we apply a similar technique as in the IC model to replace the pair-oracle with a standard oracle while maintaining $\widetilde{O}(\sqrt{T})$-regret.

poster session 5 @ blue 1, poster session 10 @ blue 1, poster session 5, poster session 10

While there has been an increasing focus on the use of game theoretic models for autonomous driving, empirical evidence shows that there are still open questions around dealing with the challenges of common knowledge assumptions as well as modeling bounded rationality. To address some of these practical challenges, we develop a framework of generalized dynamic cognitive hierarchy for both modelling naturalistic human driving behavior as well as behavior planning for autonomous vehicles (AV). This framework is built upon a rich model of level-0 behavior through the use of automata strategies, an interpretable notion of bounded rationality through safety and maneuver satisficing, and a robust response for planning. Based on evaluation on two large naturalistic datasets as well as simulation of critical traffic scenarios, we show that i) automata strategies are well suited for level-0 behavior in a dynamic level-k framework, and ii) the proposed robust response to a heterogeneous population of strategic and non-strategic reasoners can be an effective approach for game theoretic planning in AV.

poster session 5 @ blue 6, poster session 8 @ blue 6, poster session 5, poster session 8

In this paper, we propose TransMEF: a transformer-based multi-exposure image fusion framework via self-supervised multi-task learning. The framework is based on an encoder-decoder network, which can be trained on large natural datasets without the need of ground truth fusion images. We design three self-supervised reconstruction tasks according to the characteristics of multi-exposure images and conduct the three tasks simultaneously through multi-task learning, so that the network can learn the characteristics of multi-exposure images and extract more generalized features. In addition, to compensate for the defect in establishing long-range dependencies in CNN-based architectures, we design an encoder that combines a CNN module with a Transformer module to enable the network to focus on both local and global information. We evaluated our method and compared it to 11 competitive traditional and deep learning-based methods on a latest released multi-exposure image fusion benchmark dataset, and our method achieves the best performance in both subjective and objective evaluations.

poster session 5 @ red 3, poster session 12 @ red 3, poster session 5, poster session 12

Scene Graph Generation (SGG) aims to build a structured representation of a scene using objects and pairwise relationships, which benefits downstream tasks. However, current SGG methods usually suffer from sub-optimal scene graph generation because of the long-tailed distribution of training data. To address this problem, we propose Resistance Training using Prior Bias (RTPB) for the scene graph generation. Specifically, RTPB uses a distributed-based prior bias to improve models' detecting ability on less frequent relationships during training, thus improving the model generalizability on tail categories. In addition, to further explore the contextual information of objects and relationships, we design a contextual encoding backbone network, termed as Dual Transformer (DTrans). We perform extensive experiments on a very popular benchmark, VG150, to demonstrate the effectiveness of our method for the unbiased scene graph generation. In specific, our RTPB achieves an improvement of over 10\% under the mean recall when applied to current SGG methods. Furthermore, DTrans with RTPB outperforms nearly all state-of-the-art methods with a large margin.

poster session 5 @ red 3, poster session 12 @ red 3, poster session 5, poster session 12

In this paper, we propose a novel method to extend sequence-to-sequence models to accurately process sequences much longer than the ones used during training while being sample- and resource-efficient, supported by thorough experimentation. To investigate the effectiveness of our method, we apply it to the task of correcting documents already processed with Optical Character Recognition (OCR) systems using sequence-to-sequence models based on characters. We test our method on nine languages of the ICDAR 2019 competition on post-OCR text correction and achieve a new state-of-the-art performance in five of them. The strategy with the best performance involves splitting the input document in character n-grams and combining their individual corrections into the final output using a voting scheme that is equivalent to an ensemble of a large number of sequence models. We further investigate how to weigh the contributions from each one of the members of this ensemble. Our code for post-OCR correction is shared at https://github.com/jarobyte91/post_ocr_correction.

poster session 5 @ red 5, poster session 12 @ red 5, poster session 5, poster session 12

The fairness in machine learning is getting increasing attention, as its applications in different fields continue to expand and diversify. To mitigate the discriminated model behaviors between different demographic groups, we introduce a novel post-processing method to optimize over multiple fairness constraints through group-aware threshold adaptation. We propose to learn adaptive classification thresholds for each demographic group by optimizing the confusion matrix estimated from the probability distribution of a classification model output. As we only need an estimated probability distribution of model output instead of the classification model structure, our post-processing model can be applied to a wide range of classification models and improve fairness in a model-agnostic manner and ensure privacy. This even allows us to post-process existing fairness methods to further improve the trade-off between accuracy and fairness. Moreover, our model has low computational cost. We provide rigorous theoretical analysis on the convergence of our optimization algorithm and the trade-off between accuracy and fairness of our method. Our method theoretically enables a better upper bound in near optimality than existing method under same condition. Experimental results demonstrate that our method outperforms state-of-the-art methods and obtains the result that is closest to the theoretical accuracy-fairness trade-off boundary.

poster session 5 @ blue 1, poster session 10 @ blue 1, poster session 5, poster session 10

Mixed communities of organisms are found in many environments -- from the human gut to marine ecosystems -- and can have profound impact on human health and the environment. Metagenomics studies the genomic material of such communities through high-throughput sequencing that yields DNA subsequences for subsequent analysis. A fundamental problem in the standard workflow, called binning, is to discover clusters, of genomic subsequences, associated with the constituent organisms. Inherent noise in the subsequences, various biological constraints that need to be imposed on them and the skewed cluster size distribution exacerbate the difficulty of this unsupervised learning problem. In this paper, we present a new formulation using a graph where the nodes are subsequences and edges represent homophily information. In addition, we model biological constraints providing heterophilous signal about nodes that cannot be clustered together. We solve the binning problem by developing new algorithms for (i) graph representation learning that preserves both homophily relations and heterophily constraints (ii) constraint-based graph clustering method that addresses the problems of skewed cluster size distribution. Extensive experiments, on real and synthetic datasets, demonstrate that our approach, called RepBin, outperforms a wide variety of competing methods. Our constraint-based graph representation learning and clustering methods, that may be useful in other domains as well, advance the state-of-the-art in both metagenomics binning and graph representation learning.

poster session 5 @ blue 4, poster session 7 @ blue 4, oral session 7 @ blue 4, poster session 5, poster session 7, oral session 7

Underwater images suffer from degradation due to light scattering and absorption. It remains challenging to restore such degraded images using deep neural networks since real-world paired data is scarcely available while synthetic paired data cannot approximate real-world data perfectly. In this paper, we propose an UnSupervised Underwater Image Restoration method (USUIR) by leveraging the homology property between a raw underwater image and a re-degraded image. Specifically, USUIR first estimates three latent components of the raw underwater image, i.e., the global background light, the transmission map, and the scene radiance (the clean image). Then, a re-degraded image is generated by randomly mixing up the estimated scene radiance and the raw underwater image. We demonstrate that imposing a homology constraint between the raw underwater image and the re-degraded image is equivalent to minimize the restoration error, and hence can be used for the unsupervised restoration. Extensive experiments show that USUIR achieves promising performance in both inference time and restoration quality.

poster session 5 @ red 2, poster session 12 @ red 2, poster session 5, poster session 12

We present a deep neural network (DNN) for visually classifying whether a person is wearing a protective face mask. Our DNN can be deployed on a resource-limited, sub-10-cm nano-drone: this robotic platform is an ideal candidate to fly in human proximity and perform ubiquitous visual perception safely. This paper describes our pipeline, starting from the dataset collection; the selection and training of a full-precision (i.e., float32) DNN; a quantization phase (i.e., int8), enabling the DNN's deployment on a parallel ultra-low power (PULP) system-on-chip aboard our target nano-drone. Results demonstrate the efficacy of our pipeline with a mean area under the ROC curve score of 0.81, which drops by only ~2% when quantized to 8-bit for deployment.

poster session 5 @ red 2, poster session 12 @ red 2, poster session 5, poster session 12

We model the societal task of redistricting political districts as a partitioning problem: Given a set of $n$ points in the plane, each belonging to one of two parties, and a parameter $k$, our goal is to compute a partition $\Pi$ of the plane into regions so that each region contains roughly $\sigma = n/k$ points. $\Pi$ should satisfy a notion of ``local'' fairness, which is related to the notion of core, a well-studied concept in cooperative game theory. A region is associated with the majority party in that region, and a point is unhappy in $\Pi$ if it belongs to the minority party. A group $D$ of roughly $\sigma$ contiguous points is called a deviating group with respect to $\Pi$ if majority of points in $D$ are unhappy in $\Pi$. The partition $\Pi$ is locally fair if there is no deviating group with respect to $\Pi$. This paper focuses on a restricted case when points lie in $1$D. The problem is non-trivial even in this case. We consider both adversarial and ``beyond worst-case" settings for this problem. For the former, we characterize the input parameters for which a locally fair partition always exists; we also show that a locally fair partition may not exist for certain parameters. We then consider input models where there are ``runs'' of red and blue points. For such clustered inputs, we show that a locally fair partition may not exist for certain values of $\sigma$, but an approximate locally fair partition exists if we allow some regions to have smaller sizes. We finally present a polynomial-time algorithm for computing a locally fair partition if one exists.

poster session 5 @ blue 6, poster session 8 @ blue 6, poster session 5, poster session 8

Reference-based super-resolution (RefSR) has made significant progress in producing realistic textures using an external reference (Ref) image. However, existing RefSR methods obtain high-quality correspondence matchings consuming quadratic computation resources with respect to the input size, limiting its application. Moreover, these approaches usually suffer from scale misalignments between the low-resolution (LR) image and Ref image. In this paper, we propose an Accelerated Multi-Scale Aggregation network (AMSA) for Reference-based Super-Resolution, including Coarse-to-Fine Embedded PatchMatch (CFE-PatchMatch) and Multi-Scale Dynamic Aggregation (MSDA) module. To improve matching efficiency, we design a novel Embedded PatchMacth scheme with random samples propagation, which involves end-to-end training with asymptotic linear computational cost to the input size. To further reduce computational cost and speed up convergence, we apply the coarse-to-fine strategy on Embedded PatchMacth constituting CFE-PatchMatch. To fully leverage reference information across multiple scales and enhance robustness to scale misalignment, we develop the MSDA module consisting of Dynamic Aggregation and Multi-Scale Aggregation. The Dynamic Aggregation corrects minor scale misalignment by dynamically aggregating features, and the Multi-Scale Aggregation brings robustness to large scale misalignment by fusing multi-scale information. Experimental results show that the proposed AMSA achieves superior performance over state-of-the-art approaches on both quantitative and qualitative evaluations.

poster session 5 @ red 2, poster session 12 @ red 2, poster session 5, poster session 12

The one-shot multi-object tracking, which integrates object detection and ID embedding extraction into a unified network, has achieved groundbreaking results in recent years. However, current one-shot trackers solely rely on single-frame detections to predict candidate bounding boxes, which may be unreliable when facing disastrous visual degradation, e.g., motion blur, occlusions. Once a target bounding box is mistakenly classified as background by the detector, the temporal consistency of its corresponding tracklet will be no longer maintained. In this paper, we set out to restore the bounding boxes misclassified as ``fake background'' by proposing a re-check network. The re-check network innovatively expands the role of ID embedding from data association to motion forecasting by effectively propagating previous tracklets to the current frame with a small overhead. Note that the propagation results are yielded by an independent and efficient embedding search, preventing the model from over-relying on detection results. Eventually, it helps to reload the ``fake background'' and repair the broken tracklets. Building on a strong baseline CSTrack, we construct a new one-shot tracker and achieve favorable gains by 70.7 ➡ 76.4, 70.6 ➡ 76.3 MOTA on MOT16 and MOT17, respectively. It also reaches a new state-of-the-art MOTA and IDF1 performance. Code is released at https://github.com/JudasDie/SOTS.

poster session 5 @ red 1, poster session 9 @ red 1, poster session 5, poster session 9

The optimization with orthogonality has been shown useful in training deep neural networks (DNNs). To impose orthogonality on DNNs, both computational efficiency and stability are important. However, existing methods utilizing Riemannian optimization or hard constraints can only ensure stability while those using soft constraints can only improve efficiency. In this paper, we propose a novel method, named Feedback Gradient Descent (FGD), to our knowledge, the first work showing high efficiency and stability simultaneously. FGD induces orthogonality based on the simple yet indispensable Euler discretization of a continuous-time dynamical system on the tangent bundle of the Stiefel manifold. In particular, inspired by a numerical integration method on manifolds called Feedback Integrators, we propose to instantiate it on the tangent bundle of the Stiefel manifold for the first time. In our extensive image classification experiments, FGD comprehensively outperforms the existing state-of-the-art methods in terms of accuracy, efficiency, and stability.

poster session 5 @ blue 2, poster session 10 @ blue 2, poster session 5, poster session 10

Influence blocking maximization (IBM) is crucial in many critical real-world problems such as rumors prevention and epidemic containment. The existing work suffers from: (1) concentrating on uniform costs at the individual level, (2) mostly utilizing greedy approaches to approximate optimization, (3) lacking a proper graph representation for influence estimates. To address these issues, this research introduces a neural network model dubbed Neural Influence Blocking (\algo) for improved approximation and enhanced influence blocking effectiveness. The code is available at https://github.com/oates9895/NIB.

poster session 5 @ blue 5, poster session 9 @ blue 5, poster session 5, poster session 9

This paper starts by revealing a surprising finding: without any learning, a randomly initialized CNN can localize objects surprisingly well. That is, a CNN has an inductive bias to naturally focus on objects, named as Tobias ("The object is at sight") in this paper. This empirical inductive bias is further analyzed and successfully applied to self-supervised learning (SSL). A CNN is encouraged to learn representations that focus on the foreground object, by transforming every image into various versions with different backgrounds, where the foreground and background separation is guided by Tobias. Experimental results show that the proposed Tobias significantly improves downstream tasks, especially for object detection. This paper also shows that Tobias has consistent improvements on training sets of different sizes, and is more resilient to changes in image augmentations.

poster session 5 @ red 3, poster session 12 @ red 3, oral session 12 @ red 3, poster session 5, poster session 12, oral session 12

Graph Convolutional Network (GCN) has shown remarkable potential of exploring graph representation. However, the GCN aggregating mechanism fails to generalize to networks with heterophily where most nodes have neighbors from different classes, which commonly exists in real-world networks. In order to make the propagation and aggregation mechanism of GCN suitable for both homophily and heterophily (or even their mixture), we introduce block modelling into the framework of GCN so that it can realize “block-guided classified aggregation”, and automatically learn the corresponding aggregation rules for neighbors of different classes. By incorporating block modelling into the aggregation process, GCN is able to automatically aggregate information from homophilic and heterophilic neighbors discriminately according to their homophily degree. We compared our algorithm with state-of-art methods which deal with the heterophily problem. Empirical results demonstrate the superiority of our new approach over existing methods in heterophilic datasets while maintaining a competitive performance in homophilic datasets.

poster session 5 @ blue 5, poster session 9 @ blue 5, oral session 9 @ blue 5, poster session 5, poster session 9, oral session 9

Modal regression, a widely used regression protocol, has been extensively investigated in statistical and machine learning communities due to its robustness to outlier and heavy-tailed noises. Understanding modal regression's theoretical behavior can be fundamental in learning theory. Despite significant progress in characterizing its statistical property, the majority results are based on the assumption that samples are independent and identical distributed (i.i.d.), which is too restrictive for real-world applications. This paper concerns about the statistical property of regularized modal regression (RMR) within an important dependence structure - Markov dependent. Specifically, we establish the upper bound for RMR estimator under moderate conditions and give an explicit learning rate. Our results show that the Markov dependence impacts on the generalization error in the way that sample size would be discounted by a multiplicative factor depending on the spectral gap of underlying Markov chain. This result shed a new light on characterizing the theoretical underpinning for robust regression.

poster session 5 @ blue 1, poster session 10 @ blue 1, poster session 5, poster session 10

With the wide application of electronic health records (EHR) in healthcare facilities, health event prediction with deep learning has gained more and more attention. A common feature of EHR data used for deep-learning-based predictions is historical diagnoses. Existing work mainly regards a diagnosis as an independent disease and does not consider clinical relations among diseases in a visit. Many machine learning approaches assume disease representations are static in different visits of a patient. However, in real practice, multiple diseases that are frequently diagnosed at the same time reflect hidden patterns that are conducive to prognosis. Moreover, the development of a disease is not static since some diseases can emerge or disappear and show various symptoms in different visits of a patient. To effectively utilize this combinational disease information and explore the dynamics of diseases, we propose a novel context-aware learning framework using transition functions on dynamic disease graphs. Specifically, we construct a global disease co-occurrence graph with multiple node properties for disease combinations. We design dynamic subgraphs for each patient's visit to leverage global and local contexts. We further define three diagnosis roles in each visit based on the variation of node properties to model disease transition processes. Experimental results on two real-world EHR datasets show that the proposed model outperforms state of the art in predicting health events.

poster session 5 @ blue 4, poster session 7 @ blue 4, poster session 5, poster session 7

Heterogeneous Graph Neural Networks (HGNNs) have drawn increasing attention in recent years and achieved outstanding performance in many tasks. However, despite their wide use, there is currently no understanding of their robustness to adversarial attacks. In this work, we first systematically study the robustness of HGNNs and show that they can be easily fooled by adding the adversarial edge between the target node and large-degree node (i.e., hub). Furthermore, we show two key reasons for such vulnerability of HGNNs: one is perturbation enlargement effect, i.e., HGNNs, failing to encode transiting probability, will enlarge the effect of the adversarial hub in comparison of GCNs, and the other is soft attention mechanism, i.e., such mechanism assigns positive attention values to obviously unreliable neighbors. Based on the two facts, we propose a novel robust HGNN framework \textit{RoHe} against topology adversarial attacks by equipping an attention purifier, which can prune malicious neighbors based on topology and feature. Specifically, to eliminate the perturbation enlargement, we introduce the metapath-based transiting probability as the prior criterion of the purifier, restraining the confidence of malicious neighbors from the adversarial hub. Then the purifier learns to mask out neighbors with low confidence, thus can effectively alleviate the negative effect of malicious neighbors in the soft attention mechanism. Extensive experiments on different benchmark datasets for multiple HGNNs are conducted, where the considerable improvement of HGNNs under adversarial attacks will demonstrate the effectiveness and generalization ability of our defense framework.

poster session 5 @ blue 5, poster session 9 @ blue 5, poster session 5, poster session 9

Traditional subspace clustering aims to cluster data lying in a union of linear subspaces. The vectorization of high-dimensional data to 1-D vectors to perform clustering ignores much of the structure intrinsic to such data. To preserve said structure, in this work we exploit clustering in a high-order tensor space rather than a vector space. We develop a novel low-tensor-rank representation (LTRR) for unfolded matrices of tensor data lying in a low-rank tensor space. The representation coefficient matrix of an unfolding matrix is tensorized to a 3-order tensor, and the low-tensor-rank constraint is imposed on the transformed coefficient tensor to exploit the self-expressiveness property. Then, inspired by the multi-view clustering framework, we develop a multi-mode tensor space clustering algorithm (MMTSC) that can deal with tensor space clustering with or without missing entries. The tensor is unfolded along each mode, and the coefficient matrices are obtained for each unfolded matrix. The low tensor rank constraint is imposed on a tensor combined from transformed coefficient tensors of each mode, such that the proposed method can simultaneously capture the low rank property for the data within each tensor space and maintain cluster consistency across different modes. Experimental results demonstrate that the proposed MMTSC algorithm can outperform existing clustering algorithms in many cases.

poster session 5 @ blue 1, poster session 10 @ blue 1, poster session 5, poster session 10

We extend the classic regret minimization framework for approximating equilibria in matrix games by greedily weighing iterates based on the regrets observed at runtime. Theoretically, our method retains all previous convergence rate guarantees. Empirically, experiments on large random matrix games and normal-form subgames of the AI benchmark Diplomacy show that greedy weights outperforms previous methods whenever sampling is used, sometimes by several orders of magnitude.

poster session 5 @ blue 6, poster session 8 @ blue 6, poster session 5, poster session 8

We consider the problem of reforming an envy-free matching when each agent is assigned a single item. Given an envy-free matching, we consider an operation to exchange the item of an agent with an unassigned item preferred by the agent that results in another envy-free matching. We repeat this operation as long as we can. We prove that the resulting envy-free matching is uniquely determined up to the choice of an initial envy-free matching, and can be found in polynomial time. We call the resulting matching a reformist envy-free matching, and then we study a shortest sequence to obtain the reformist envy-free matching from an initial envy-free matching. We prove that a shortest sequence is computationally hard to obtain even when each agent accepts at most four items and each item is accepted by at most three agents. On the other hand, we give polynomial-time algorithms when each agent accepts at most three items or each item is accepted by at most two agents. Inapproximability and fixed-parameter (in)tractability are also discussed.

poster session 5 @ blue 6, poster session 8 @ blue 6, poster session 5, poster session 8

Readers of novels need to identify and learn about the characters as they develop an understanding of the plot. The paper presents an end-to-end automated pipeline for literary character identification and ongoing work for extracting and comparing character representations for full-length English novels. The character identification pipeline involves a named entity recognition (NER) module with F1 score of 0.85, a coreference resolution module with F1 score of 0.76, and a disambiguation module using both heuristic and algorithmic approaches. Ongoing work compares event extraction as well as speech extraction pipelines for literary characters representations with case studies. The paper is the first to my knowledge that combines a modular pipeline for automated character identification, representation extraction and comparisons for full-length English novels.

poster session 1 @ red 5, poster session 5 @ red 5, poster session 1, poster session 5

Bayesian neural networks (BNNs) have drawn extensive interest due to the unique probabilistic representation framework. However, Bayesian neural networks have limited publicized deployments because of the relatively poor model performance in real-world applications. In this paper, we argue that the randomness of sampling in Bayesian neural networks causes errors in the updating of model parameters during training and some sampled models with poor performance in testing. To solve this, we propose to train Bayesian neural networks with Adversarial Distribution as a theoretical solution. To avoid the difficulty of calculating Adversarial Distribution analytically, we further present the Adversarial Sampling method as an approximation in practice. We conduct extensive experiments with multiple network structures on different datasets, e.g., CIFAR-10 and CIFAR-100. Experimental results validate the correctness of the theoretical analysis and the effectiveness of the Adversarial Sampling on improving model performance. Additionally, models trained with Adversarial Sampling still keep their ability to model uncertainties and perform better when predictions are retained according to the uncertainties, which further verifies the generality of the Adversarial Sampling approach.

poster session 5 @ blue 1, poster session 10 @ blue 1, poster session 5, poster session 10

Graph Neural Networks (GNNs) have shown advantages in various graph-based applications. Most existing GNNs assume strong homophily of graph structure and apply permutation-invariant local aggregation of neighbors to learn a representation for each node. However, they fail to generalize to heterophilic graphs, where most neighboring nodes have different labels or features, and the relevant nodes are distant. Few recent studies attempt to address this problem by combining multiple hops of hidden representations of central nodes (i.e., multi-hop-based approaches) or sorting the neighboring nodes based on attention scores (i.e., ranking-based approaches). As a result, these approaches have some apparent limitations. On the one hand, multi-hop-based approaches do not explicitly distinguish relevant nodes from a large number of multi-hop neighborhoods, leading to a severe over-smoothing problem. On the other hand, ranking-based models do not joint-optimize node ranking with end tasks and result in sub-optimal solutions. In this work, we present Graph Pointer Neural Networks (GPNN) to tackle the challenges mentioned above. We leverage a pointer network to select the most relevant nodes from a large amount of multi-hop neighborhoods, which constructs an ordered sequence according to the relationship with the central node. 1D convolution is then applied to extract high-level features from the node sequence. The pointer-network-based ranker in GPNN is joint-optimized with other parts in an end-to-end manner. Extensive experiments are conducted on six public node classification datasets with heterophilic graphs. The results show that GPNN significantly improves the classification performance of state-of-the-art methods. In addition, analyses also reveal the privilege of the proposed GPNN in filtering out irrelevant neighbors and reducing over-smoothing.

poster session 5 @ blue 3, poster session 10 @ blue 3, oral session 5 @ blue 3, poster session 5, poster session 10, oral session 5

Most prior work on online matching problems has been with the flexibility of keeping some vertices unmatched. We study three related online matching problems with the constraint of matching every vertex, i.e., with no rejections. We adopt a model in which vertices arrive in a uniformly random order and the edge-weights are arbitrary positive numbers. For the capacitated online bipartite matching problem in which the vertices of one side of the graph are offline and those of the other side arrive online, we give a 4.62-competitive algorithm when the capacity of each offline vertex is 2. For the online general (non-bipartite) matching problem, where all vertices arrive online, we give a 3.34-competitive algorithm. We also study the online roommate matching problem (Huzhang et al. 2017) in which each room (offline vertex) holds 2 persons (online vertices). Persons derive non-negative additive utilities from their room as well as roommate. In this model, with the goal of maximizing the sum of utilities of all persons, we give a 7.96-competitive algorithm. This is an improvement over the 24.72 approximation factor in (Huzhang et al. 2017).

poster session 5 @ blue 6, poster session 8 @ blue 6, poster session 5, poster session 8

Communication efficiency is crucial for federated learning (FL). Conducting local training steps in clients to reduce the communication frequency between clients and the server is a common method to address this issue. However, it leads to the client drift problem due to \textit{non-i.i.d.} data distributions in different clients which severely deteriorates the performance. In this work, we propose a new method to improve the training performance in cross-silo FL via maintaining double momentum buffers. One momentum buffer tracks the server model updating direction, and the other tracks the local model updating direction. Moreover, we introduce a novel momentum fusion technique to coordinate the server and local momentum buffers. We also provide the first theoretical convergence analysis involving both the server and local standard momentum SGD. Extensive deep FL experimental results show a better training performance than FedAvg and existing standard momentum SGD variants.

poster session 5 @ blue 3, poster session 10 @ blue 3, oral session 10 @ blue 3, poster session 5, poster session 10, oral session 10

For personalized ranking models, the well-calibrated probability of an item being preferred by a user has great practical value. While existing work shows promising results in image classification, probability calibration has not been much explored for personalized ranking. In this paper, we aim to estimate the calibrated probability of how likely a user will prefer an item. We investigate various parametric distributions and propose two parametric calibration methods, namely Gaussian calibration and Gamma calibration. Each proposed method can be seen as a post-processing function that maps the ranking scores of pre-trained models to well-calibrated preference probabilities, without affecting the recommendation performance. We also design the unbiased empirical risk minimization framework that guides the calibration methods to learning of true preference probability from the biased user-item interaction dataset. Extensive evaluations with various personalized ranking models on real-world datasets show that both the proposed calibration methods and the unbiased empirical risk minimization significantly improve the calibration performance.

poster session 5 @ blue 5, poster session 9 @ blue 5, oral session 5 @ blue 5, poster session 5, poster session 9, oral session 5

Hierarchical SGD (H-SGD) has emerged as a new distributed SGD algorithm for multi-level communication networks. In H-SGD, before each global aggregation, workers send their updated local models to local servers for aggregations. Despite recent research efforts, the effect of local aggregation on global convergence still lacks theoretical understanding. In this work, we first introduce a new notion of "upward" and "downward" divergences. We then use it to conduct a novel analysis to obtain a worst-case convergence upper bound for two-level H-SGD with non-IID data, non-convex objective function, and stochastic gradient. By extending this result to the case with random grouping, we observe that this convergence upper bound of H-SGD is between the upper bounds of two single-level local SGD settings, with the number of local iterations equal to the local and global update periods in H-SGD, respectively. We refer to this as the "sandwich behavior". Furthermore, we extend our analytical approach based on "upward" and "downward" divergences to study the convergence for the general case of H-SGD with more than two levels, where the "sandwich behavior" still holds. Our theoretical results provide key insights of why local aggregation can be beneficial in improving the convergence of H-SGD.

poster session 5 @ blue 3, poster session 10 @ blue 3, poster session 5, poster session 10

Source-Free Unsupervised Domain Adaptation~(SFUDA) aims to adapt a pre-trained source model to an unlabeled target domain without access to the original labeled source domain samples. Many existing SFUDA approaches apply the self-training strategy, \ie, iteratively selecting confidently predicted target samples as pseudo-labeled samples to train the model to fit the target domain. However, the self-training strategy may also suffer from the sample selection bias and the label noise of the pseudo-labeled samples. In this work, we provide a rigorous theoretical analysis on how these two issues affect the model generalization ability when applying self-training strategy for the SFUDA problem. Benefited from the theoretical analysis, we then propose a new Denoised Maximum Classifier Discrepancy (D-MCD) for SFUDA to effectively address these two issues. In particular, we first minimize the distribution mismatch between the selected pseudo-labeled samples and the rest target domain samples to alleviate the sample selection bias. Besides, we design a strong-weak self-training paradigm to denoise the selected pseudo-labeled samples, where the strong network is used to select pseudo-labeled samples while the weak network helps the strong network to filter out hard samples to avoid incorrect labels. In this way, we are able to ensure both the quality of pseudo-label and the generalization ability of the trained model on the target domain. We achieve state-of-the-art results on three domain adaptation benchmark datasets, which clearly validates the effectiveness of our proposed approach.

poster session 5 @ blue 3, poster session 10 @ blue 3, oral session 5 @ blue 3, poster session 5, poster session 10, oral session 5

Graph Convolutional Networks (GCNs) have emerged as the state-of-the-art deep learning model for representation learning on graphs. However, it remains notoriously challenging to train and inference GCNs over large graph datasets, limiting their application to large real-world graphs and hindering the exploration of deeper and more sophisticated GCN graphs. This is because as the graph size grows, the sheer number of node features and the large adjacency matrix can easily explode the required memory and data movements. To tackle the aforementioned challenges, we explore the possibility of drawing lottery tickets when sparsifying GCN graphs, i.e., subgraphs that largely shrink the adjacency matrix yet are capable of achieving accuracy comparable to or even better than their full graphs. Specifically, we for the first time discover the existence of graph early-bird (GEB) tickets that emerge at the very early stage when sparsifying GCN graphs, and propose a simple yet effective detector to automatically identify the emergence of such GEB tickets. Furthermore, we advocate graph-model co-optimization and develop a generic efficient GCN early-bird training framework dubbed GEBT that can significantly boost the efficiency of GCN training by (1) drawing joint early-bird tickets between the GCN graphs and models and (2) enabling simultaneously sparsification of both the GCN graphs and models. Experiments on various GCN models and datasets consistently validate our GEB finding and the effectiveness of our GEBT, e.g., our GEBT achieves up to 80.2% ~ 85.6% and 84.6% ~ 87.5% savings of GCN training and inference costs while offering a comparable or even better accuracy as compared to state-of-the-art methods. Our source code and supplementary appendix are available at https://github.com/RICE-EIC/Early-Bird-GCN.

poster session 5 @ blue 3, poster session 10 @ blue 3, poster session 5, poster session 10

Data augmentation has been an important ingredient for boosting performances of learned models. Prior data augmentation methods for few-shot text classification have led to great performance boosts. However, they have not been designed to capture the intricate compositional structure of natural language. As a result, they fail to generate samples with plausible and diverse sentence structures. Motivated by this, we present the data Augmentation using Lexicalized Probabilistic context-free grammars (ALP) that generates augmented samples with diverse syntactic structures with plausible grammar. The lexicalized PCFG parse trees consider both the constituents and dependencies to produce a syntactic frame that maximizes a variety of word choices in a syntactically preservable manner without specific domain experts. Experiments on few-shot text classification tasks demonstrate that ALP enhances many state-of-the-art classification methods. As a second contribution, we delve into the train-val splitting methodologies when a data augmentation method comes into play. We argue empirically that the traditional splitting of training and validation sets is sub-optimal compared to our novel augmentation-based splitting strategies that further expand the training split with the same number of labeled data. Taken together, our contributions on the data augmentation strategies yield a strong training recipe for few-shot text classification tasks.

poster session 5 @ red 4, poster session 10 @ red 4, poster session 5, poster session 10

Pansharpening is a critical yet challenging low-level vision task that aims to obtain a higher-resolution image by fusing a multispectral (MS) image and a panchromatic (PAN) image. While most pansharpening methods are based on convolutional neural network (CNN) architectures with standard convolution operations, few attempts have been made with context-adaptive/dynamic convolution, which delivers impressive results on high-level vision tasks. In this paper, we propose a novel strategy to generate local-context adaptive (LCA) convolution kernels and introduce a new global harmonic (GH) bias mechanism, exploiting image local specificity as well as integrating global information, dubbed LAGConv. The proposed LAGConv can replace the standard convolution that is context-agnostic to fully perceive the particularity of each pixel for the task of remote sensing pansharpening. Furthermore, by applying the LAGConv, we provide an image fusion network architecture, which is more effective than conventional CNN-based pansharpening approaches. The superiority of the proposed method is demonstrated by extensive experiments implemented on a wide range of datasets compared with state-of-the-art pansharpening methods. Besides, more discussions testify that the proposed LAGConv outperforms recent adaptive convolution techniques for pansharpening.

poster session 5 @ red 2, poster session 12 @ red 2, poster session 5, poster session 12

Submodular functions are at the core of many machine learning and data mining tasks. The underlying submodular functions for many of these tasks are decomposable, i.e., they are sum of several simple submodular functions. In many data intensive applications, however, the number of underlying submodular functions in the original function is so large that we need prohibitively large amount of time to process it and/or it does not even fit in the main memory. To overcome this issue, we introduce the notion of sparsification for decomposable submodular functions whose objective is to obtain an accurate approximation of the original function that is a (weighted) sum of only a few submodular functions. Our main result is a polynomial-time randomized sparsification algorithm such that the expected number of functions used in the output is independent of the number of underlying submodular functions in the original function. We also study the effectiveness of our algorithm under various constraints such as matroid and cardinality constraints. We complement our theoretical analysis with an empirical study of the performance of our algorithm.

poster session 5 @ blue 2, poster session 10 @ blue 2, poster session 5, poster session 10

Social media, blogs, and online articles are instant sources of news for internet users globally. But due to their unmoderated nature, a significant percentage of these texts are fake news or rumors. Their deceptive nature and ability to propagate instantly can have an adverse effect on society. In this work, we hypothesize that legitimacy of news has a correlation with its emotion, and propose a multi-task framework predicting both the emotion and legitimacy of news. Experimental results verify that our multi-task models outperform their single-task counterparts in terms of accuracy.

poster session 5 @ blue 4, poster session 12 @ blue 4, poster session 5, poster session 12

By interpreting Graph Neural Networks (GNNs) as the message passing from the spatial perspective, their success is attributed to Laplacian smoothing. However, it also leads to serious over-smoothing issue by stacking many layers. Recently, many efforts have been paid to overcome this issue in semi-supervised learning. Unfortunately, it is more serious in unsupervised node representation learning task due to the lack of supervision information. Thus, most of the unsupervised or self-supervised GNNs often employ \textit{one-layer GCN} as the encoder. Essentially, the over-smoothing issue is caused by the over-simplification of the existing message passing, which possesses two intrinsic limits: blind message and uniform passing. In this paper, a novel Diverse and Interactive Message Passing (DIMP) is proposed for self-supervised learning by overcoming these limits. Firstly, to prevent the message from blindness and make it interactive between two connected nodes, the message is determined by both the two connected nodes instead of the attributes of one node. Secondly, to prevent the passing from uniformness and make it diverse over different attribute channels, different propagation weights are assigned to different elements in the message. To this end, a natural implementation of the message in DIMP is the element-wise product of the representations of two connected nodes. From the perspective of numerical optimization, the proposed DIMP is equivalent to performing an overlapping community detection via expectation-maximization (EM). Both the objective function of the community detection and the convergence of EM algorithm guarantee that DMIP can prevent from over-smoothing issue. Extensive evaluations on node-level and graph-level tasks demonstrate the superiority of DIMP on improving performance and overcoming over-smoothing issue.

poster session 5 @ blue 5, poster session 9 @ blue 5, poster session 5, poster session 9

Deep and large pre-trained language models (e.g., BERT, GPT-3) are state-of-the-art for various natural language processing tasks. However, the huge size of these models brings challenges to fine-tuning and online deployment due to latency and cost constraints. Existing knowledge distillation methods reduce the model size, but they may encounter difficulties transferring knowledge from the teacher model to the student model due to the limited data from the downstream tasks. In this work, we propose AD^2, a novel and effective data augmentation approach to improve the task-specific knowledge transfer when compressing large pre-trained transformer models. Different from prior methods, AD^2 performs distillation by using an enhanced training set that contains both original inputs and adversarially perturbed samples that mimic the output distribution from the teacher. Experimental results show that this method allows better transfer of knowledge from the teacher to the student during distillation, producing student models that retain 99.6\% accuracy of the teacher model while outperforming existing task-specific knowledge distillation baselines by 1.2 points on average over a variety of natural language understanding tasks. Moreover, compared with alternative data augmentation methods, such as text-editing-based approaches, AD^2 is up to 28 times faster while achieving comparable or higher accuracy. In addition, when AD^2 is combined with more advanced task-agnostic distillation, we can advance the state-of-the-art performance even more. On top of the encouraging performance, this paper also provides thorough ablation studies and analysis. The discovered interplay between KD and adversarial data augmentation for compressing pre-trained Transformers may further inspire more advanced KD algorithms for compressing even larger scale models.

poster session 5 @ blue 3, poster session 10 @ blue 3, poster session 5, poster session 10

Recently, deep learning has been proven to be a promising approach in standard dynamic range (SDR) image compression. However, due to the wide luminance distribution of high dynamic range (HDR) images and the lack of large standard datasets, developing a deep model for HDR image compression is much more challenging. To tackle this issue, we view HDR data as distributional shifts of SDR data and the HDR image compression can be modeled as an out-of-distribution generalization (OoD) problem. Herein, we propose a novel out-of-distribution (OoD) HDR image compression framework (OoDHDR-codec). It learns the general representation across HDR and SDR environments, and allows the model to be trained effectively using a large set of SDR datases supplemented with much fewer HDR samples. Specifically, OoDHDR-codec consists of two branches to process the data from two environments. The SDR branch is a standard blackbox network. For the HDR branch, we develop a hybrid system that models luminance masking and tone mapping with white-box modules and performs content compression with black-box neural networks. To improve the generalization from SDR training data on HDR data, we introduce an invariance regularization term to learn the common representation for both SDR and HDR compression. Extensive experimental results show that the OoDHDR codec achieves strong competitive in-distribution performance and state-of-the-art OoD performance. To the best of our knowledge, our proposed approach is the first work to model HDR compression as OoD generalization problems and our OoD generalization algorithmic framework can be applied to any deep compression model in addition to the network architectural choice demonstrated in the paper.

poster session 5 @ blue 5, poster session 9 @ blue 5, oral session 5 @ blue 5, poster session 5, poster session 9, oral session 5

Multi-task learning has been widely used in real-world recommenders to predict different types of user feedback. Most prior works focus on designing network architectures for bottom layers as a means to share the knowledge about input features representations. However, since they adopt task-specific binary labels as supervised signals for training, the knowledge about how to accurately rank items is not fully shared across tasks. In this paper, we aim to enhance knowledge transfer for multi-task personalized recommendat optimization objectives. We propose a Cross-Task Knowledge Distillation (CrossDistil) framework in recommendation, which consists of three procedures. 1) Task Augmentation: We introduce auxiliary tasks with quadruplet loss functions to capture cross-task fine-grained ranking information, which could avoid task conflicts by preserving the cross-task consistent knowledge; 2) Knowledge Distillation: We design a knowledge distillation approach based on augmented tasks for sharing ranking knowledge, where tasks' predictions are aligned with a calibration process; 3) Model Training: Teacher and student models are trained in an end-to-end manner, with a novel error correction mechanism to speed up model training and improve knowledge quality. Comprehensive experiments on a public dataset and our production dataset are carried out to verify the effectiveness of TKD as well as the necessity of its key components.

poster session 5 @ blue 5, poster session 9 @ blue 5, poster session 5, poster session 9

When dealing with a series of imminent issues, humans can naturally concentrate on a subset of these concerning issues by prioritizing them according to their contributions to motivational indices, e.g., the probability of winning a game. This idea of concentration offers insights into reinforcement learning of sophisticated Large-scale Multi-Agent Systems (LMAS) participated by hundreds of agents. In such an LMAS, each agent receives a long series of entity observations at each step, which can overwhelm existing aggregation networks such as graph attention networks and cause inefficiency. In this paper, we propose a concentration network called ConcNet. First, ConcNet scores the observed entities considering several motivational indices, e.g., expected survival time and state value of the agents, and then ranks, prunes, and aggregates the encodings of observed entities to extract features. Second, distinct from the well-known attention mechanism, ConcNet has a unique motivational subnetwork to explicitly consider the motivational indices when scoring the observed entities. Furthermore, we present a concentration policy gradient architecture that can learn effective policies in LMAS from scratch. Extensive experiments demonstrate that the presented architecture has excellent scalability and flexibility, and significantly outperforms existing methods on LMAS benchmarks.

poster session 5 @ red 3, poster session 12 @ red 3, oral session 5 @ red 3, poster session 5, poster session 12, oral session 5

Explainable machine learning (ML) has gained traction in recent years due to the increasing adoption of ML-based systems in many sectors. Algorithmic Recourses (ARs) provide "what if" feedback of the form "if an input datapoint were x' instead of x, then an ML-based system's output would be y' instead of y." Recourses are attractive due to their actionable feedback, amenability to existing legal frameworks, and fidelity to the underlying ML model. Yet, current recourse approaches are single shot---that is, they assume x can change to x' in a single time period. We propose a novel stochastic-control-based approach that generates sequential recourses, that is, recourses that allow x to move stochastically and sequentially across intermediate states to a final state x'. Our approach is model agnostic and black box. Furthermore, the calculation of recourses is amortized such that once trained, it applies to multiple datapoints without the need for re-optimization. In addition to these primary characteristics, our approach admits optional desiderata such as adherence to the data manifold, respect for causal relations, and sparsity---identified by past research as desirable properties of recourses. We evaluate our approach using three real-world datasets and show successful generation of sequential recourses that respect other counterfactual desiderata.

poster session 5 @ blue 1, poster session 10 @ blue 1, poster session 5, poster session 10

In social choice theory, (Kemeny) rank aggregation is a well-studied problem where the goal is to combine rankings from multiple voters into a single ranking on the same set of items. Since rankings can reveal preferences of voters (which a voter might like to keep private), it is important to aggregate preferences in such a way to preserve privacy. In this work, we present differentially private algorithms for rank aggregation in the pure and approximate settings along with distribution-independent utility upper and lower bounds. In addition to bounds in the central model, we also present utility bounds for the local model of differential privacy.

poster session 5 @ blue 2, poster session 10 @ blue 2, poster session 5, poster session 10

Dynamic Vision Sensor (DVS) can asynchronously output the events reflecting apparent motion of objects with microsecond resolution, and shows great application potential in monitoring and other fields. However, the output event stream of existing DVS inevitably contains background activity noise (BA noise) due to dark current and junction leakage current, which will affect the temporal correlation of objects, resulting in deteriorated motion estimation performance. Particularly, the existing filter-based denoising methods cannot be directly applied to suppress the noise in event stream, since there is no spatial correlation. To address this issue, this paper presents a novel progressive framework, in which a Motion Estimation (ME) module and an Event Denoising (ED) module are jointly optimized in a mutually reinforced manner. Specifically, based on the maximum sharpness criterion, ME module divides the input event into several segments by adaptive clustering in a motion compensating warp field, and captures the temporal correlation of event stream according to the clustered motion parameters. Taking temporal correlation as guidance, ED module calculates the confidence that each event belongs to real activity events, and transmits it to ME module to update energy function of motion segmentation for noise suppression. The two steps are iteratively updated until stable motion segmentation results are obtained. Extensive experimental results on both synthetic and real datasets demonstrate the superiority of our proposed approaches against the State-Of-The-Art (SOTA) methods.

poster session 5 @ red 2, poster session 12 @ red 2, poster session 5, poster session 12

Unsupervised domain adaptation has recently emerged as an effective paradigm for generalizing deep neural networks to new target domains. However, there is still enormous potential to be tapped to reach the fully supervised performance. In this paper, we present a novel active learning strategy to assist knowledge transfer in the target domain, dubbed active domain adaptation. We start from an observation that energy-based models exhibit free energy biases when training (source) and test (target) data come from different distributions. Inspired by this inherent mechanism, we empirically reveal that a simple yet efficient energy-based sampling strategy sheds light on selecting the most valuable target samples than existing approaches requiring particular architectures or computation of the distances. Our algorithm, Energy-based Active Domain Adaptation (EADA), queries groups of targe data that incorporate both domain characteristic and instance uncertainty into every selection round. Meanwhile, by aligning the free energy of target data compact around the source domain via a regularization term, domain gap can be implicitly diminished. Through extensive experiments, we show that EADA surpasses state-of-the-art methods on well-known challenging benchmarks with substantial improvements, making it a useful option in the open world. Code is available at https://github.com/BIT-DA/EADA.

poster session 5 @ blue 3, poster session 10 @ blue 3, poster session 5, poster session 10

Optimal transport (OT) plays an essential role in various areas like machine learning and deep learning. However, computing discrete optimal transport plan for large scale problems with adequate accuracy and efficiency is still highly challenging. Recently, methods based on the Sinkhorn algorithm add an entropy regularizer to the prime problem and get a trade off between efficiency and accuracy. In this paper, we propose a novel algorithm to further improve the efficiency and accuracy based on Nesterov's smoothing technique. Basically, the non-smooth c-transform of the Kantorovich potential is approximated by the smooth Log-Sum-Exp function, which finally smooths the original non-smooth Kantorovich dual functional. The smooth Kantorovich functional can be optimized by the fast proximal gradient algorithm (FISTA) efficiently. Theoretically, the computational complexity of the proposed method is given by $O(n^{\frac{5}{2}} \sqrt{\log n} /\epsilon)$, which is lower than current estimation of the Sinkhorn algorithm. Empirically, compared with the Sinkhorn algorithm, our experimental results demonstrate that the proposed method achieves faster convergence and better accuracy with the same parameter.

poster session 5 @ blue 1, poster session 10 @ blue 1, poster session 5, poster session 10

Few-Shot Learning (FSL) is a challenging task, i.e., how to recognize novel classes with few examples? Pre-training based methods effectively tackle the problem by pre-training a feature extractor and then predicting novel classes via a cosine nearest neighbor classifier with mean-based prototypes. Nevertheless, due to the data scarcity, the mean-based prototypes are usually biased. In this paper, we attempt to diminish the prototype bias by regarding it as a prototype optimization problem. To this end, we propose a novel meta-learning based prototype optimization framework to rectify prototypes, i.e., introducing a meta-optimizer to optimize prototypes. Although the existing meta-optimizers can also be adapted to our framework, they all overlook a crucial gradient bias issue, i.e., the mean-based gradient estimation is also biased on sparse data. To address the issue, we regard the gradient and its flow as meta-knowledge and then propose a novel Neural Ordinary Differential Equation (ODE)-based meta-optimizer to polish prototypes, called MetaNODE. In this meta-optimizer, we first view the mean-based prototypes as initial prototypes, and then model the process of prototype optimization as continuous-time dynamics specified by a Neural ODE. A gradient flow inference network is carefully designed to learn to estimate the continuous gradient flow for prototype dynamics. Finally, the optimal prototypes can be obtained by solving the Neural ODE. Extensive experiments on miniImagenet, tieredImagenet, and CUB-200-2011 show the effectiveness of our method.

poster session 5 @ red 4, poster session 10 @ red 4, poster session 5, poster session 10

Acquiring the most representative examples via active learning (AL) can benefit many data-dependent computer vision tasks by minimizing efforts of image-level or pixel-wise annotations. In this paper, we propose a novel Collaborative Panoptic-Regional Active Learning framework (CPRAL) to address the semantic segmentation task. For a small batch of images initially sampled with pixel-wise annotations, we employ panoptic information to initially select unlabeled samples. Considering the class imbalance in the segmentation dataset, we import a Regional Gaussian Attention module (RGA) to achieve semantics-biased selection. The subset is highlighted by vote entropy and then attended by Gaussian kernels to maximize the biased regions. We also propose a Contextual Labels Extension (CLE) to boost regional annotations with contextual attention guidance. With the collaboration of semantics-agnostic panoptic matching and region-biased selection and extension, our CPRAL can strike a balance between labeling efforts and performance and compromise the semantics distribution. We perform extensive experiments on Cityscapes and BDD10K datasets and show that CPRAL outperforms the cutting-edge methods with impressive results and less labeling proportion.

poster session 5 @ blue 3, poster session 10 @ blue 3, poster session 5, poster session 10

Existing CNNs-Based RGB-D salient object detection (SOD) networks are all required to be pretrained on the ImageNet to learn the hierarchy features which helps provide a good initialization. However, the collection and annotation of large-scale datasets are time-consuming and expensive. In this paper, we utilize self-supervised representation learning (SSL) to design two pretext tasks: the cross-modal auto-encoder and the depth-contour estimation. Our pretext tasks require only a few and unlabeled RGB-D datasets to perform pretraining, which makes the network capture rich semantic contexts and reduce the gap between two modalities, thereby providing an effective initialization for the downstream task. In addition, for the inherent problem of cross-modal fusion in RGB-D SOD, we propose a consistency-difference aggregation (CDA) module that splits a single feature fusion into multi-path fusion to achieve an adequate perception of consistent and differential information. The CDA module is general and suitable for cross-modal and cross-level feature fusion. Extensive experiments on six benchmark datasets show that our self-supervised pretrained model performs favorably against most state-of-the-art methods pretrained on ImageNet.

poster session 5 @ red 2, poster session 12 @ red 2, poster session 5, poster session 12

We formalize a voting model for plurality elections that combines Iterative Voting and Calculus of Voting. Each iteration, autonomous agents simultaneously maximize the utility they expect from candidates. Agents are aware of neither other individuals’ preferences or choices, nor of the distribution preferences come from. They know only of candidates’ current expected vote shares (e.g. from a poll, network, etc) and with that calculate expected rewards from each candidate, pondering the probability that voting for each would alter the final outcome. We define the general form of those pivotal probabilities, then we derive both efficient exact calculations as well as heuristics. Lastly, we both prove formally that the model converges with asymptotically large electorates and show empirically (via simulations) that it nearly always converges even with very few agents.

poster session 5 @ blue 6, poster session 8 @ blue 6, poster session 5, poster session 8

It is an essential product requirement of Yahoo Mail to distinguish between personal and machine-generated emails. The old production classifier in Yahoo Mail was based on a simple logistic regression model. That model was trained by aggregating features at the SMTP address level. We propose building deep models at the message level. We built and trained four individual CNN models: (1) a content model with subject and content as input; (2) a sender model with sender email address and name as input; (3) an action model by analyzing email recipients’ action patterns and correspondingly generating target labels based on senders’ opening/deleting behaviors; (4) a salutation model by utilizing senders’ “explicit salutation”signal as positive labels. Next, we built a final full model after exploring different combinations of the above four models. Experimental results on editorial data show that our full model improves the adjusted recall from 70.5% to 78.8% compared to the old production model, while at the same time lifts the precision from 94.7% to 96.0%. Our full model also significantly beats the state-of-the-art Bert model at this task. This full model has been deployed into the current production system (Yahoo Mail 6).

poster session 5 @ blue 2, poster session 10 @ blue 2, poster session 5, poster session 10

Syntax has been shown to benefit Coreference Resolution from incorporating long-range dependencies and structured information captured by syntax trees, either in traditional statistical machine learning based systems or recently proposed neural models. However, most leading systems use only dependency trees. We argue that constituent trees also encode important information, such as explicit span-boundary signals captured by nested multi-word phrases, extra linguistic labels and hierarchical structures useful for detecting anaphora. In this work, we propose a simple yet effective graph-based method to incorporate constituent syntactic structures. Moreover, we also explore to utilise higher-order neighbourhood information to encode rich structures in constituent trees. A novel message propagation mechanism is therefore proposed to enable information flow among elements in syntax trees. Experiments on the English and Chinese portions of OntoNotes 5.0 benchmark show that our proposed model either beats a strong baseline or achieves new state-of-the-art performance.

poster session 5 @ red 5, poster session 12 @ red 5, poster session 5, poster session 12

RGBT tracking usually suffers from various challenge factors, such as fast motion, scale variation, illumination variation, thermal crossover and occlusion, to name a few. Existing works often study fusion models to solve all challenges simultaneously, and it requires fusion models complex enough and training data large enough, which are usually difficult to be constructed in real-world scenarios. In this work, we disentangle the fusion process via the challenge attributes, and thus propose a novel Attribute-based Progressive Fusion Network (APFNet) to increase the fusion capacity with a small number of parameters while reducing the dependence on large-scale training data. In particular, we design five attribute-specific fusion branches to integrate RGB and thermal features under the challenges of thermal crossover, illumination variation, scale variation, occlusion and fast motion respectively. By disentangling the fusion process, we can use a small number of parameters for each branch to achieve robust fusion of different modalities and train each branch using the small training subset with the corresponding attribute annotation. Then, to adaptive fuse features of all branches, we design an aggregation fusion module based on SKNet. Finally, we also design an enhancement fusion transformer to strengthen the aggregated feature and modality-specific features. Experimental results on benchmark datasets demonstrate the effectiveness of our APFNet against other state-of-the-art methods.

poster session 5 @ red 1, poster session 9 @ red 1, poster session 5, poster session 9

Reinforcement learning typically assumes that agents observe feedback for their actions immediately, but in many real-world applications (like recommendation systems) feedback is observed in delay. This paper studies online learning in episodic Markov decision processes (MDPs) with unknown transitions, adversarially changing costs and unrestricted delayed feedback. That is, the costs and trajectory of episode $k$ are revealed to the learner only in the end of episode $k + d^k$, where the delays $d^k$ are neither identical nor bounded, and are chosen by an oblivious adversary. We present novel algorithms based on policy optimization that achieve near-optimal high-probability regret of $\sqrt{K + D}$ under full-information feedback, where $K$ is the number of episodes and $D = \sum_{k} d^k$ is the total delay. Under bandit feedback, we prove similar $\sqrt{K + D}$ regret assuming the costs are stochastic, and $(K + D)^{2/3}$ regret in the general case. We are the first to consider regret minimization in the important setting of MDPs with delayed feedback.

poster session 5 @ blue 1, poster session 10 @ blue 1, poster session 5, poster session 10

Student assessment is one of the most fundamental tasks in the field of AI Education (AIEd). One of the most common approach to student assessment is Knowledge Tracing (KT), which evaluates a student's knowledge state by predicting whether the student will answer a given question correctly or not. However, in the context of multiple choice (polytomous) questions, conventional KT approaches are limited in that they only consider the binary (dichotomous) correctness label (i.e., correct or incorrect), and disregard the specific option chosen by the student. Meanwhile, Option Tracing (OT) attempts to model a student by predicting which option they will choose for a given question, but overlooks the correctness information. In this paper, we propose Dichotomous-Polytomous Multi-Task Learning (DP-MTL), a multi-task learning framework that combines KT and OT for more precise student assessment. In particular, we show that the KT objective acts as a regularization term for OT in the DP-MTL framework, and propose an appropriate architecture for applying our method on top of existing deep learning-based KT models. We experimentally confirm that DP-MTL significantly improves both KT and OT performances, and also benefits downstream tasks such as Score Prediction (SP).

poster session 5 @ blue 4, poster session 7 @ blue 4, poster session 5, poster session 7

Detecting depression based on non-verbal behaviors has received great attention. However, most prior work on detecting depression mainly focused on detecting depressed individuals in laboratory settings, which are difficult to be generalized in practice. In addition, little attention has been paid to analyzing the non-verbal behaviors of depressed individuals in the wild. Therefore, in this paper, we present a multimodal depression dataset, D-Vlog, which consists of 961 vlogs (i.e., around 160 hours) collected from YouTube, which can be utilized in developing depression detection models based on the non-verbal behavior of individuals in real-world scenario. We develop a multimodal deep learning model that uses acoustic and visual features extracted from collected data to detect depression. Our proposed model employs the cross-attention mechanism to effectively capture the relationship across acoustic and visual features, and generates useful multimodal representations for depression detection. The extensive experimental results demonstrate that the proposed model significantly outperforms other baseline models. We believe our dataset and the proposed model are useful for analyzing and detecting depressed individuals based on non-verbal behavior.

poster session 5 @ red 6, poster session 12 @ red 6, poster session 5, poster session 12

Waves in the oceans are one of the greatest sources of renewable energy and are a promising resource to tackle climate challenges by decarbonizing energy generation. Lowering the Levelized Cost of Energy (LCOE) for energy generation from ocean waves is key for competitiveness with other forms of clean energy like wind and solar. This requires complex controllers to maximize efficiency for state-of-the-art multi-generator industrial Wave Energy Converters (WEC), which optimizes the reactive forces of the generators on multiple legs of WEC. This paper intro-duces Multi-Agent Reinforcement Learning controller (MARL) architectures which can handle these multiple objectives for LCOE, helping the increase in energy capture efficiency boosting revenue, reducing structural stress to limit maintenance and operating cost, and adaptively and proactively protect the wave energy converter from catastrophic weather events, preserving investments and lowering effective capital cost. These architectures include 2-agent and 3-agent MARL implementing proximal policy optimization (PPO) with various optimizations to help sustain the training convergence in the complex hyperplane without falling off the cliff. The design for trust is implemented to assure the operation of WEC within a safe zone of mechanical compliance and guarantee mechanical integrity. This is achieved through reward shaping for multiple objectives of energy capture and penalty for harmful motions to minimize stress and lower the cost of maintenance. We achieved double-digit gains in energy capture efficiency across the waves of different principal frequencies over the baseline Spring Damper controller with the proposed MARL controllers.

poster session 2 @ red 6, poster session 5 @ red 6, poster session 2, poster session 5

Citing and describing related literature are crucial to scientific writing. Many existing approaches show encouraging performance in citation recommendation, but are unable to accomplish the more challenging and onerous task of citation text generation. In this paper, we propose a novel disentangled representation based model DisenCite to automatically generate the citation text through integrating paper text and citation graph. A key novelty of our method compared with existing approaches is to generate context-specific citation text, empowering the generation of different types of citations for the same paper. In particular, we first build and make available a graph enhanced contextual citation dataset (GCite) with 25K edges in different types characterized by citation contained sections over 4.8K research papers. Based on this dataset, we encode each paper according to both textual contexts and structure information in the heterogeneous citation graph. The resulted paper representations are then disentangled by the mutual information regularization between this paper and its neighbors in graph. Extensive experiments demonstrate the superior performance of our method comparing to state-of-the-art approaches. We further conduct ablation and case studies to reassure that the improvement of our method comes from generating the context-specific citation through incorporating the citation graph.

poster session 5 @ red 5, poster session 12 @ red 5, oral session 5 @ red 5, poster session 5, poster session 12, oral session 5

Modern recommendation systems are mostly based on implicit feedback data which can be quite noisy due to false positives (FPs) caused by many reasons, such as misclicks or quick curiosity. Numerous recommendation algorithms based on collaborative filtering have leveraged post-click user behavior (e.g., skip) to identify false positives. They effectively involved these false positives in the model supervision as negative-like signals. Yet, false positives had not been considered in existing session-based recommendation systems (SBRs) although they provide just as deleterious effects. To resolve false positives in SBRs, we first introduce FP-Metric model which reformulates the objective of the session-based recommendation with FP constraints into metric learning regularization. In addition, we propose FP-AdaMetric that enhances the metric-learning regularization terms with an adaptive module that elaborately calculates the impact of FPs inside sequential patterns. We verify that FP-AdaMetric improves several session-based recommendation models' performances in terms of Hit Rate (HR), MRR, and NDCG on datasets from different domains including music, movie, and game. Furthermore, we show that the adaptive module plays a much more crucial role in FP-AdaMetric model than in other baselines.

poster session 5 @ blue 5, poster session 9 @ blue 5, poster session 5, poster session 9

RGB-infrared person re-identification is an emerging cross-modality re-identification task, which is very challenging due to significant modality discrepancy between RGB and infrared images. In this work, we propose a novel modality-adaptive mixup and invariant decomposition (MID) approach for RGB-infrared person re-identification towards learning modality-invariant and discriminative representations. MID designs a modality-adaptive mixup scheme to generate suitable mixed modality images between RGB and infrared images for mitigating the inherent modality discrepancy at the pixel-level. It formulates modality mixup procedure as Markov decision process, where an actor-critic agent learns dynamical and local linear interpolation policy between different regions of cross-modality images under a deep reinforcement learning framework. Such policy guarantees modality-invariance in a more continuous latent space and avoids manifold intrusion by the corrupted mixed modality samples. Moreover, to further counter modality discrepancy and enforce invariant visual semantics at the feature-level, MID employs modality-adaptive convolution decomposition to disassemble a regular convolution layer into modality-specific basis layers and a modality-shared coefficient layer. Extensive experimental results on two challenging benchmarks demonstrate superior performance of MID over state-of-the-art methods.

poster session 5 @ red 3, poster session 12 @ red 3, poster session 5, poster session 12

We propose an end-to-end image compression and analysis model with Transformers, targeting to the cloud-based image classification application. Instead of placing an existing Transformer-based image classification model directly after an image codec, we aim to redesign the Vision Transformer (ViT) model to perform image classification from the compressed features and facilitate image compression with the long-term information from the Transformer. Specifically, we first replace the patchify stem (i.e., image splitting and embedding) of the ViT model with a lightweight image encoder modelled by a convolutional neural network. The compressed features generated by the image encoder are injected convolutional inductive bias and are fed to the Transformer for image classification bypassing image reconstruction. Meanwhile, we propose a feature aggregation module to fuse the compressed features with the selected intermediate features of the Transformer, and feed the aggregated features to a deconvolutional neural network for image reconstruction. The aggregated features can obtain the long-term information from the self-attention mechanism of the Transformer and improve the compression performance. The rate-distortion-accuracy optimization problem is finally solved by a two-step training strategy. Experimental results demonstrate the effectiveness of the proposed model in both the image compression and the classification tasks.

poster session 5 @ red 2, poster session 12 @ red 2, oral session 5 @ red 2, poster session 5, poster session 12, oral session 5

Traditionally, clustered federated learning groups clients with the same data distribution into a cluster, so that every client is uniquely associated with one data distribution and helps train a model for this distribution. We relax this hard association assumption to soft clustered federated learning, which allows every local dataset to follow a mixture of multiple source distributions. We propose FedSoft, which trains both locally personalized models and high-quality cluster models in this setting. FedSoft limits client workload by using proximal updates to require the completion of only one optimization task from a subset of clients in every communication round. We show, analytically and empirically, that FedSoft effectively exploits similarities between the source distributions to learn personalized and cluster models that perform well.

poster session 5 @ blue 3, poster session 10 @ blue 3, oral session 10 @ blue 3, poster session 5, poster session 10, oral session 10

As a key characteristic in audio-visual speech recognition (AVSR), relating linguistic information observed across visual and audio data has been a challenge, benefiting not only audio/visual speech recognition (ASR/VSR) but also for manipulating data within/across modalities. In this paper, we present a feature disentanglement-based framework for jointly addressing the above tasks. By advancing cross-modal mutual learning strategies, our model is able to convert visual or audio-based linguistic features into modality-agnostic representations. Such derived linguistic representations not only allow one to perform ASR, VSR, and AVSR, but also to manipulate audio and visual data output based on the desirable subject identity and linguistic content information. We perform extensive experiments on different recognition and synthesis tasks to show that our model performs favorably against state-of-the-art approaches on each individual task, while ours is a unified solution that is able to jointly tackle the aforementioned audio-visual learning tasks.

poster session 5 @ red 3, poster session 12 @ red 3, oral session 12 @ red 3, poster session 5, poster session 12, oral session 12

Assessing the impact of the COVID-19 crisis on economies is fundamental to tailor the responses of the governments to recover from the crisis. In this paper, we present a novel approach to assessing the economic impact with a large-scale credit card transaction dataset at a fine granularity. For this purpose, we develop a fine-grained economic-epidemiological modeling framework COVID-EENet, which is featured with a two-level deep neural network. In support of the fine-grained EEM, COVID-EENet learns the impact of nearby mass infection cases on the changes of local economies in each district. Through the experiments using the nationwide dataset, given a set of active mass infection cases, COVID-EENet is shown to precisely predict the sales changes in two or four weeks for each district and business category. Therefore, policymakers can be informed of the predictive impact to put in the most effective mitigation measures. Overall, we believe that our work opens a new perspective of using financial data to recover from the economic crisis. For public use in this urgent problem, we release the source code at https://github.com/kaist-dmlab/COVID-EENet.

poster session 5 @ red 6, poster session 12 @ red 6, oral session 5 @ red 6, poster session 5, poster session 12, oral session 5

We consider mechanisms for truthfully eliciting probabilistic predictions from a group of experts. The standard approach -- using a proper scoring rule to separately reward each expert -- is not robust to collusion: experts may collude to misreport their beliefs in a way that guarantees them a larger total reward no matter the eventual outcome. Chun and Shachter (2011) termed any such collusion "arbitrage" and asked whether there is any truthful elicitation mechanism that makes arbitrage impossible. We resolve this question positively, exhibiting a class of strictly proper arbitrage-free contract functions. These contract functions have two parts: one ensures that the total reward of a coalition of experts depends only on the average of their reports; the other ensures that changing this average report hurts the experts under at least one outcome.

poster session 5 @ blue 6, poster session 8 @ blue 6, oral session 5 @ blue 6, poster session 5, poster session 8, oral session 5

In recent years, self-supervised representation learning for skeleton-based action recognition has been developed with the advance of contrastive learning methods. The existing contrastive learning methods use normal augmentations to construct similar positive samples, which limits the ability to explore novel movement patterns. In this paper, to make better use of the movement patterns introduced by extreme augmentations, a Contrastive Learning framework utilizing Abundant Information Mining for self-supervised action Representation (AimCLR) is proposed. First, the extreme augmentations and the Energy-based Attention-guided Drop Module (EADM) are proposed to obtain diverse positive samples, which bring novel movement patterns to improve the universality of the learned representations. Second, since directly using extreme augmentations may not be able to boost the performance due to the drastic changes in original identity, the Dual Distributional Divergence Minimization Loss (D3M Loss) is proposed to minimize the distribution divergence in a more gentle way. Third, the Nearest Neighbors Mining (NNM) is proposed to further expand positive samples to make the abundant information mining process more reasonable. Exhaustive experiments on NTU RGB+D 60, PKU-MMD, NTU RGB+D 120 datasets have verified that our AimCLR can significantly perform favorably against state-of-the-art methods under a variety of evaluation protocols with observed higher quality action representations. Our code is available at https://github.com/Levigty/AimCLR.

poster session 5 @ red 1, poster session 9 @ red 1, oral session 5 @ red 1, poster session 5, poster session 9, oral session 5

Emotion detection from face is an important problem and has received attention from industry and academia. Although emotion recognition from videos has a very high performance, emotion recognition from a single image stays a challenging task. In this paper, we try to use information from videos to do emotion recognition on a single image. More specifically, we leverage contrastive loss for pretraining the network on the videos and experiment with different sampling methods to select consistently hard triplets for continual learning of the network. Once the embeddings have been trained, we test them on a standard emotion classification task. Our method significantly improves the performance of the models and shows the efficacy of self-supervision in emotion recognition.

poster session 5 @ red 1, poster session 9 @ red 1, poster session 5, poster session 9

Wearable devices that integrate multiple sensors, processors, and communication technologies have the potential to transform mobile health for remote monitoring of health parameters. However, the small form factor of the wearable devices limits the battery size and operating lifetime. As a result, the devices require frequent recharging, which has limited their widespread adoption. Energy harvesting has emerged as an effective method towards sustainable operation of wearable devices. Unfortunately, energy harvesting alone is not sufficient to fulfill the energy requirements of wearable devices. This paper studies the novel problem of adaptive energy management towards the goal of self-sustainable wearables by using harvested energy to supplement the battery energy and to reduce manual recharging by users. To solve this problem, we propose a principled algorithm referred as AdaEM. There are two key ideas behind AdaEM. First, it uses machine learning (ML) methods to learn predictive models of user activity and energy usage patterns. These models allow us to estimate the potential of energy harvesting in a day as a function of the user activities. Second, it reasons about the uncertainty in predictions and estimations from the ML models to optimize the energy management decisions using a dynamic robust optimization (DyRO) formulation. We propose a light-weight solution for DyRO to meet the practical needs of deployment. We validate the AdaEM approach on a wearable device prototype consisting of solar and motion energy harvesting using real-world data of user activities. Experiments show that AdaEM achieves solutions that are within 5% of the optimal with less than 0.005% execution time and energy overhead.

poster session 2 @ red 6, poster session 5 @ red 6, poster session 2, poster session 5

The daily practice of sharing images on social media raises a severe issue about privacy leakage. To address the issue, privacy-leaking image detection is studied recently, with the goal to automatically identify images that may leak privacy. Recent advance on this task benefits from focusing on crucial objects via pretrained object detectors and modeling their correlation. However, these methods have two limitations: 1) they neglect other important elements like scenes, textures, and objects beyond the capacity of pretrained object detectors. 2) the correlation among objects is fixed, but a fixed correlation is not appropriate for all the images. To overcome the limitations, we propose the Dynamic Region-Aware Graph Convolutional Network (DRAG) that dynamically finds out crucial regions including objects and other important elements, and model their correlation adaptively for each input image. To find out crucial regions, we cluster spatially-correlated feature channels into several region-aware feature maps. Furthermore, we dynamically model the correlation with the self-attention mechanism and explore the interaction among the regions with a graph convolutional network. The DRAG achieved an accuracy of 87% on the largest dataset for privacy-leaking image detection, which is 10 percentage points higher than the state of the art. The further case study demonstrates that it found out crucial regions containing not only objects but other important elements like textures. The code and more details are in https://github.com/guang-yanng/DRAG.

poster session 5 @ red 6, poster session 12 @ red 6, oral session 5 @ red 6, poster session 5, poster session 12, oral session 5

Deep learning models have excelled in solving many problems in Natural Language Processing, but are susceptible to extensive vulnerabilities. We offer a solution to this vulnerability by using random perturbations such as spelling correction, synonym substitution, or dropping the word. These perturbations are applied to random words in random sentences to defend NLP models against adversarial attacks. Our defense methods are successful in returning attacked models to their original accuracy within statistical significance.

poster session 1 @ red 5, poster session 5 @ red 5, poster session 1, poster session 5

Market popularity prediction has always been a hot research topic, such as sales prediction and crowdfunding prediction. Most of these studies put the perspective on isolated markets, relying on the knowledge of certain market to maximize the prediction performance. However, these market-specific approaches are restricted by the knowledge limitation of isolated markets and incapable of the complicated and potential relations among different markets, especially some with strong dependence such as the financing market and sales market. Fortunately, we discover potentially symbiotic relations between the financing market and the sales market, which provides us with an opportunity to co-promote the popularity predictions of both markets. Thus, for bridgly learning the knowledge interactions between financing market and sales market, we propose a cross-market approach, namely CATN: Cooperative-competitive Attention Transfer Network, which could effectively transfer knowledge of financing capability from the crowdfunding market and sales prospect from the E-commerce market. Specifically, for capturing the complicated relations especially the cooperation or complement of items and enhancing the knowledge transfer between the two heterogeneous markets, we design a novel Cooperative Attention; meanwhile, for finely computing the relations of items especially the competition in specific same market, we further design Competitive Attentions for the two markets respectively. Besides, we also distinguish aligned features and unique features to adapt the cross-market predictions. With the real-world datasets collected from Indiegogo and Amazon, we construct extensive experiments on three types of datasets from the two markets and the results demonstrate the effectiveness and generalization of our CATN model.

poster session 5 @ blue 4, poster session 7 @ blue 4, poster session 5, poster session 7

In this article, we proposed a technique for action localization and recognition from long untrimmed videos. It consists of C3D CNN model followed by the action mining using the localization model, where the KNN classifier is used. We segment the video into expressible sub-action known as action-bytes. The pseudo labels have been used to train the localization model, which makes the trimmed videos untrimmed for action-bytes. We present experimental results on the recent benchmark trimmed video dataset “Thumos14”.

poster session 5 @ red 1, poster session 9 @ red 1, poster session 5, poster session 9

$k$-center is one of the most popular clustering models. While it admits a simple 2-approximation in polynomial time in general metrics, the Euclidean version is NP-hard to approximate within a factor of 1.93, even in the plane, if one insists the dependence on $k$ in the running time be polynomial. Without this restriction, a classic algorithm yields a $2^{O((k\log k)/{\epsilon})}dn$-time $(1+\epsilon)$-approximation for Euclidean $k$-center, where $d$ is the dimension. In this work, we give a faster algorithm for small dimensions: roughly speaking an $O^*(2^{O((1/\epsilon)^{O(d)} \cdot k^{1-1/d} \cdot \log k)})$-time $(1+\epsilon)$-approximation. In particular, the running time is roughly $O^*(2^{O((1/\epsilon)^{O(1)}\sqrt{k}\log k)})$ in the plane. We complement our algorithmic result with a matching hardness lower bound. We also consider a well-studied generalization of $k$-center, called Non-uniform $k$-center (NUkC), where we allow different radii clusters. NUkC is NP-hard to approximate within any factor, even in the Euclidean case. We design a $2^{O(k\log k)}n^2$ time $3$-approximation for NUkC, and a $2^{O((k\log k)/\epsilon)}dn$ time $(1+\epsilon)$-approximation for Euclidean NUkC. The latter time bound matches the bound for $k$-center.

poster session 5 @ blue 1, poster session 10 @ blue 1, poster session 5, poster session 10

High-resolution daytime satellite imagery has become a promising source to study economic activities. These images display detailed terrain over large areas and allow zooming into smaller neighborhoods. Existing methods, however, have utilized images only in a single-level geographical unit. This research presents a deep learning model to predict economic indicators via aggregating traits observed from multiple levels of geographical units. The model first measures hyperlocal economy over small communities via ordinal regression. The next step extracts district-level features by summarizing interconnection among hyperlocal economies. In the final step, the model estimates economic indicators of districts via aggregating the hyperlocal and district information. Our new multi-level learning model substantially outperforms strong baselines in predicting key indicators such as population, purchasing power, and energy consumption. The model is also robust against data shortage; the trained features from one country can generalize to other countries when evaluated with data gathered from Malaysia, the Philippines, Thailand, and Vietnam. We discuss the multi-level model's implications for measuring inequality, which is the essential first step in policy and social science research on inequality and poverty.

poster session 5 @ red 6, poster session 12 @ red 6, poster session 5, poster session 12

This paper studies learning node representations with graph neural networks (GNNs) for unsupervised scenario. Specifically, we derive a theoretical analysis and provide an empirical demonstration about the non-steady performance of GNNs over different graph datasets, when the supervision signals are not appropriately defined. The performance of GNNs depends on both the node feature smoothness and the locality of graph structure. To smooth the discrepancy of node proximity measured by graph topology and node feature, we proposed SAIL - a novel self-augmented graph contrastive learning framework, with two complementary self-distilling regularization modules, i.e., intra- and inter-graph knowledge distillation. We demonstrate the competitive performance of SAIL on a variety of graph applications. Even with a single GNN layer, SAIL has consistently competitive or even better performance on various benchmark datasets, comparing with state-of-the-art baselines.

poster session 5 @ blue 5, poster session 9 @ blue 5, poster session 5, poster session 9

State-of-the-art deep NLP models have achieved impressive improvements on many tasks. However, they are found to be vulnerable to some perturbations. Before they are widely adopted, the fundamental issues of robustness need to be addressed. In this paper, we design a robustness enhancement method to defend against word substitution perturbation, whose basic idea is to fight perturbation with perturbation. We find that: although many well-trained deep models are not robust in the setting of the presence of adversarial samples, they satisfy weak robustness. That means they can handle most non-crafted perturbations well. Taking advantage of the weak robustness property of deep models, we utilize non-crafted perturbations to resist the adversarial perturbations crafted by attackers. Our method contains two main stages. The first stage is using randomized perturbation to conform the input to the data distribution. The second stage is using randomized perturbation to eliminate the instability of prediction results and enhance the robustness guarantee. Experimental results show that our method can significantly improve the ability of deep models to resist the state-of-the-art adversarial attacks while maintaining the prediction performance on the original clean data.

poster session 5 @ red 5, poster session 12 @ red 5, poster session 5, poster session 12

Sub-seasonal forecasting (SSF) is the prediction of key climate variables such as temperature and precipitation on the 2-week to 2-month time horizon. Skillful SSF would have substantial societal value in areas such as agricultural productivity, hydrology and water resource management, and emergency planning for extreme events such as droughts and wildfires. Despite its societal importance, SSF has stayed a challenging problem compared to both short-term weather forecasting and long-term seasonal forecasting. Recent studies have shown the potential of machine learning (ML) models to advance SSF. In this paper, for the first time, we perform a fine-grained comparison of a suite of modern ML models with start-of-the-art physics-based dynamical models from the Subseasonal Experiment (SubX) project for SSF in the western contiguous United States. Additionally, we explore mechanisms to enhance the ML models by using forecasts from dynamical models. Empirical results illustrate that, on average, ML models outperform dynamical models while the ML models tend to generate forecasts with conservative magnitude compared to the SubX models. Further, we illustrate that ML models make forecasting errors under extreme weather conditions, e.g., cold waves due to the polar vortex, highlighting the need for separate models for extreme events. Finally, we show that suitably incorporating dynamical model forecasts as inputs to ML models can substantially improve the forecasting performance of the ML models. The SSF dataset constructed for the work and code for the ML models are released along with the paper for the benefit of the artificial intelligence community.

poster session 5 @ blue 4, poster session 7 @ blue 4, poster session 5, poster session 7

We develop new adaptive algorithms for variational inequalities with monotone operators, which capture many problems of interest, notably convex optimization and convex-concave saddle point problems. Our algorithms automatically adapt to unknown problem parameters such as the smoothness and the norm of the operator, and the variance of the stochastic evaluation oracle. We show that our algorithms are universal and simultaneously achieve the optimal convergence rates in the non-smooth, smooth, and stochastic settings. The convergence guarantees of our algorithms improve over existing adaptive methods by a $\Omega(\sqrt{\ln T})$ factor, matching the optimal non-adaptive algorithms. Additionally, prior works require that the optimization domain is bounded. In this work, we remove this restriction and give algorithms for unbounded domains that are adaptive and universal. Our general proof techniques can be used for many variants of the algorithm using one or two operator evaluations per iteration. The classical methods based on the ExtraGradient/MirrorProx algorithm require two operator evaluations per iteration, which is the dominant factor in the running time in many settings.

poster session 5 @ blue 2, poster session 10 @ blue 2, oral session 10 @ blue 2, poster session 5, poster session 10, oral session 10

Deep learning has made remarkable achievements for single image haze removal. However, existing deep dehazing models only give deterministic results without discussing the uncertainty of them. There exist two types of uncertainty in the dehazing models: aleatoric uncertainty that comes from noise inherent in the observations and epistemic uncertainty that accounts for uncertainty in the model. In this paper, we propose a novel uncertainty-driven dehazing network (UDN) that improves the dehazing results by exploiting the relationship between the uncertain and confident representations. We first introduce an Uncertainty Estimation Block (UEB) to predict the aleatoric and epistemic uncertainty together. Then, we propose an Uncertainty-aware Feature Modulation (UFM) block to adaptively enhance the learned features. UFM predicts a convolution kernel and channel-wise modulation cofficients conitioned on the uncertainty weighted representation. Moreover, we develop an uncertainty-driven self-distillation loss to improve the uncertain representation by transferring the knowledge from the confident one. Extensive experimental results on synthetic datasets and real-world images show that UDN achieves significant quantitative and qualitative improvements, outperforming the state-of-the-arts.

poster session 5 @ red 2, poster session 12 @ red 2, oral session 5 @ red 2, poster session 5, poster session 12, oral session 5

Implicit discourse relation recognition (IDRR) is a challenging but crucial task in discourse analysis. Most existing methods train multiple models to predict multi-level labels independently, while ignoring the dependence between hierarchically structured labels. In this paper, we consider multi-level IDRR as a conditional label sequence generation task and propose a Label Dependence-aware Sequence Generation Model (LDSGM) for it. Specifically, we first design a label attentive encoder to learn the global representation of an input instance and its level-specific contexts, where the label dependence is integrated to obtain better label embeddings. Then, we employ a label sequence decoder to output the predicted labels in a top-down manner, where the predicted higher-level labels are directly used to guide the label prediction at the current level. We further develop a mutual learning enhanced training method to exploit the label dependence in a bottom-up direction, which is captured by an auxiliary decoder introduced during training. Experimental results on the PDTB dataset show that our model achieves the state-of-the-art performance on multi-level IDRR. We release our code at https://github.com/nlpersECJTU/LDSGM.

poster session 5 @ red 5, poster session 12 @ red 5, poster session 5, poster session 12

Convolutional neural networks (CNNs) have been commonly applied in the area of the Electroencephalography (EEG)-based Motor Imagery (MI) classification, significantly pushing the boundary of the state-of-the-art. In order to simultaneously decode the discriminative features and eliminate the negative effects of non-Gaussian noise and outliers in the motor imagery data, in this abstract, we propose a novel robust supervision signal, called Correntropy based Center Loss (CCL), for CNN training, which utilizes the correntropy induced distance as the objective measure. It is encouraging to see that the CNN model trained by the combination of softmax loss and CCL loss outperforms the state-of-the-art models on two public datasets.

poster session 5 @ red 1, poster session 9 @ red 1, poster session 5, poster session 9